Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Cell barrier function of resident peritoneal macrophages in post-operative adhesions

Cell barrier function of resident peritoneal macrophages in post-operative adhesions ARTICLE https://doi.org/10.1038/s41467-021-22536-y OPEN Cell barrier function of resident peritoneal macrophages in post-operative adhesions 1 1 1 1 1 1 Tomoya Ito , Yusuke Shintani , Laura Fields , Manabu Shiraishi , Mihai‑Nicolae Podaru , Satoshi Kainuma , 1 1 1 1 1 Kizuku Yamashita , Kazuya Kobayashi , Mauro Perretti , Fiona Lewis-McDougall & Ken Suzuki Post-operative adhesions are a leading cause of abdominal surgery-associated morbidity. Exposed fibrin clots on the damaged peritoneum, in which the mesothelial barrier is dis- rupted, readily adhere to surrounding tissues, resulting in adhesion formation. Here we show High − that resident F4/80 CD206 peritoneal macrophages promptly accumulate on the lesion and form a ‘macrophage barrier’ to shield fibrin clots in place of the lost mesothelium in mice. Depletion of this macrophage subset or blockage of CD11b impairs the macrophage barrier and exacerbates adhesions. The macrophage barrier is usually insufficient to fully preclude the adhesion formation; however, it could be augmented by IL-4-based treatment or adoptive transfer of this macrophage subset, resulting in robust prevention of adhesions. By contrast, monocyte-derived recruited peritoneal macrophages are not involved in the macrophage barrier. These results highlight a previously unidentified cell barrier function of a specific macrophage subset, also proposing an innovative approach to prevent post-operative adhesions. William Harvey Research Institute, Barts and The London School of Medicine and Dentistry, Queen Mary University of London, London, UK. email: t.ito@qmul.ac.uk; ken.suzuki@qmul.ac.uk NATURE COMMUNICATIONS | (2021) 12:2232 | https://doi.org/10.1038/s41467-021-22536-y | www.nature.com/naturecommunications 1 1234567890():,; ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-22536-y 21–25 ost-operative adhesions are pathological connections peritoneal cavity, remains contentious . For instance, clo- binding organs or tissues across a virtual space, e.g., the dronate liposome-induced depletion of peritoneal macrophages Pperitoneal, pericardial or pleural cavity. These are some of attenuates the intra-abdominal adhesion formation in a gauze- the leading causes of post-operative morbidity, resulting in var- induced adhesion model in mice , suggesting a causative role of ious complications . Abdominal adhesions occur in 79–93% of macrophages. On the other hand, adoptive transfer of peritoneal patients who have undergone a major abdominal or pelvic macrophages reduces the adhesion formation in a parietal peri- 2,3 21 procedure . These can cause pain and distress and may result in toneal muscular defect model in rabbits , indicating an anti- more critical complications, i.e., bowel obstruction and female adhesion effect of macrophages. Although the different species infertility . The only effective treatment for post-operative and/or models used in these studies may be a reason for the 5 26 adhesions, once developed, is corrective surgery , whereas this contradicting observations , the heterogeneity and diversity of procedure is invasive and often leads to a recurrence of peritoneal macrophages may also explain this inconsistency. 6 High − adhesions , suggesting the importance of the prevention. The Here we show that resident F4/80 CD206 peritoneal Low + most common preventive approach for post-operative abdominal macrophages, but not recruited F4/80 CD206 macrophages, adhesions is currently implantation or administration of bioma- have an ability to form a “cell barrier” that determines the for- terial barrier products, including artificial films, liquid, or gel, mation of post-operative abdominal adhesions in a mouse model. which keep the injured tissue-surface separated from the neigh- This previously unidentified macrophage barrier is formed via boring tissue/organ . These barrier products are, however, effec- CD11b, but usually insufficient to fully shield the exposed fibrin tive in only half of the patients and have limitations in their clots, thereby allowing the adhesion formation. However, we 8–10 practicability . Therefore, there is a great need for a more describe that IL-4-based treatment augments the macrophage effective and more widely applicable preventative treatment for barrier and results in robust prevention of post-operative adhe- post-operative abdominal adhesions. To this aim, it is essential to sions, proposing an innovative anti-adhesion strategy that is further understand the cellular and molecular mechanism based on a distinctive concept from the conventional treatments. underpinning the adhesion formation post-surgery. The major initiating factor for post-operative adhesion for- Results mation is the production of fibrin clots . Following a surgery- Abdominal adhesion formation within a day of surgery.We associated injury, the coagulation cascade is activated, leading to first investigated the local cellular dynamics related to the the formation of fibrin clots on the surface of damaged tissues . abdominal adhesion formation using a mouse model based on The adhesive polymerization sites on the exposed fibrin clot ischemic button creation on the parietal peritoneum (Supple- 13,14 promote the surrounding tissue to bind with them . The mentary Fig. 1a). The severity of adhesions was evaluated by the endogenous fibrinolysis ability, in which plasmin and plasmino- 4,28 reported adhesion-scoring system . This model demonstrated gen activators play a role, is not so substantial as to attenuate the that the adhesions between the ischemic button and surrounding fibrin clot formation or adhesion formation . Previous studies tissues were formed as early as day 1 post-surgery (Supplemen- have suggested an involvement of various types of cells in this tary Fig. 1b, c). Histological assessments using hematoxylin and 16,17 initial process for post-operative adhesion formation , eosin staining revealed that the mesothelial layer on the ischemic 18–20 including activated mesothelial cells and neutrophils . How- button surface was disrupted by day 1 and that there was an ever, the role of macrophages, which are a major cell type in the accumulation of nucleated cells, which have different Naive (Day 0) Day 1 Day 3 Day 5 (B) (B) (B) (A) (A) (A) (A) Intestine Intestine Intestine Zoom (A) Zoom (A) Zoom (A) Zoom (A) Zoom (B) Zoom (B) Zoom (B) Fig. 1 Post-operative adhesions are formed as early as 1 day after surgery in the mouse ischemic button model. Representative images of hematoxylin- eosin staining of the ischemic button at different time points. Zooms (A) and (B) show the non-adhesion area (A) and adhesion area (B) of the ischemic button, respectively. The inserted images in Zooms (A) present higher magnification views of the black dashed boxes. The naive peritoneum was used as control (Day 0). n = 3 independent experiments. Scale bars represent 2 mm in the lower magnification images and 100 μm in the higher magnification images. 2 NATURE COMMUNICATIONS | (2021) 12:2232 | https://doi.org/10.1038/s41467-021-22536-y | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-22536-y ARTICLE ab Naive (Day 0) Day 1 Day 5 Ischemic button (Day 1) DAPI PDPN F4/80 Peritoneum DAPI PDPN F4/80 Ischemic Peritoneal cavity button Zoom Remote area c d Naive (Day 0) Day 1 Day 3 Day 5 F4/80 cells Peritoneum DAPI PDPN Peritoneal cavity PDPN cells DAPI Fibrin Days after surgery DAPI PDPN F4/80 Fibrin DAPI Button F4/80 DAPI PDPN Intestine F4/80 Fibrin Zoom Zoom Fig. 2 F4/80 macrophages accumulate on fibrin clots in the damaged peritoneum. a Representative tiling image of immunofluorescence staining for the ischemic button on day 1 post-surgery. Cross-sections were stained for PDPN and F4/80. The inserted images present the higher magnification of the surface area of the ischemic button and the remote area from the ischemic button. n = 3 independent ischemic buttons. Scale bar, 1 mm. b Whole-mount immunohistostaining of the ischemic button surface. n = 3 independent ischemic buttons. Scale bars, 100 μm. c, d Representative immunofluorescence images and quantification of ischemic buttons created in the mouse peritoneal membrane. Cross-sections were stained for PDPN, F4/80, and fibrin. The naive peritoneum was examined as “Day 0”. The graph shows the time course of the cover ratio of the ischemic button surface by PDPN mesothelial cells or F4/80 macrophages. n = 6 (Day 0), 5 (Day 1), 7 (Day 3), and 6 (Day 5) mice. Data represent the mean ± SEM. Scale bar, 100 μm. e Representative immunofluorescence images of the adhesion formed between the ischemic button and surrounding tissue (intestine) on day 3 post-surgery. n = 3 independent experiments. Scale bars, 200 μm. morphologies from naive mesothelial cells, on the button surface button was mostly covered by a monolayer of F4/80 macro- (Fig. 1). At the sites of adhesion formation, the mesothelial layer phages instead of PDPN cells (Fig. 2a). We then further pro- was lost by day 1, and connective tissue with cell infiltration was gressed the analysis of every ischemic button by using both formed to bind the ischemic button and the surrounding tissue, whole-mount and cross-section immunohistofluorescent staining. i.e., the intestine. Before the surgical insult, a monolayer of PDPN mesothelial cells with a cobblestone-like morphology completely covered + − the parietal peritoneum, while the majority of these PDPN cells F4/80 CD206 macrophages accumulate on fibrin clots.To disappeared from the button surface by day 1 post-surgery identify the accumulated cells on the ischemic button surface, we (Fig. 2b–d). This disappearance of mesothelial cells was further first performed immunohistofluorescent staining for F4/80 and confirmed by immunostaining for two other mesothelial podoplanin (PDPN) using an ischemic button sample that was markers , cytokeratin 19 and Mesothelin (Supplementary free from adhesion on day 1 after surgery. The result exhibited a Fig. 2a, b). The disrupted mesothelial cell layer was, however, remarkable change in the peritoneal surface of the ischemic reconstructed with PDPN cells by day 5 (Fig. 2b–d). It was button. The parietal peritoneum adjacent and remote to the observed that fibrin clots were formed on the surface of the ischemic button was entirely shielded by the PDPN mesothelial ischemic button with a peak at day 3 (Fig. 2c). Of note, F4/80 cell monolayer, while by contrast the peritoneal surface of the NATURE COMMUNICATIONS | (2021) 12:2232 | https://doi.org/10.1038/s41467-021-22536-y | www.nature.com/naturecommunications 3 Cover ratio of button surface (%) ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-22536-y a b Naive (Day 0) 4 hrs Day 1 Day 3 Day 5 High - 38.6% 3.30% 0.09% 7.99% 22.8% 10 0.6 F4/80 CD206 10 0.4 0.2 0.85% 1.04% 3.04% 8.61% 5.84% Low + F4/80 CD206 01 23 45 3 4 3 4 3 4 3 4 3 4 Days after surgery 010 10 010 10 010 10 010 10 010 10 Anti-CD206-Alexa488 13.0 11.4 c d e ± 0.3 ± 0.2 Day 3 Day 3 Day 5 -6 P = 1.06 x 10 5 *** Ischemic button creation 10 4 High - 10 F4/80 CD206 Low + F4/80 CD206 Liposome DID i.p. injection Low (Day 2 after surgery) F4/80 CD206 + 5 F4/80 CD206 - - F4/80 CD206 Analysis for peritoneal cavity cells 3 4 3 4 5 3 4 5 010 10 010 10 10 010 10 10 (Day 3 or 5 after surgery) Anti-CD206-Alexa488 Liposome DID Fig. 3 Two subsets of peritoneal macrophages exhibit different cellular properties and distinct dynamics post-surgery. a, b Representative flow High − Low + cytometry plots (a) and cell counts (b) of F4/80 CD206 macrophages (red) and F4/80 CD206 (gray) macrophages in the peritoneal fluid after ischemic button creation in mice. n = 7 (Naive; Day 0), 6 (4 h), 8 (Day 1), 8 (Day 3) and 9 (Day 5) mice. See Supplementary Fig. 5 for the gating strategy. Data are shown as the mean ± SEM. c Schematic of the phagocytosis uptake assay. d Representative flow cytometry histogram showing the phagocytosis High − Low + Low − activity of peritoneal F4/80 CD206 macrophages (red) and F4/80 CD206 (gray) macrophages in comparison with F4/80 CD206 (orange), − + − − F4/80 CD206 (green) and F4/80 CD206 (blue) cells in the peritoneal cavity. n = 3 independent experiments. e Diameter of sorted peritoneal F4/ High − Low + 80 CD206 and F4/80 CD206 macrophages. Data are presented as box-and-whisker plots (interquartile ranges (IQRs) as boxes, with the median High − as a black line and the whiskers extending up to the most extreme points within 1.5-fold IQR). n = 109 (F4/80 CD206 ) cells and 154 (F4/ Low + 80 CD206 ) cells. ***P < 0.001, two-tailed Student’s t-test. High − macrophages accumulated on these lesions and appeared to activity (Fig. 3c, d). Of note, while F4/80 CD206 macro- substitute for the disrupted mesothelial barrier albeit in part phages were dominant (~40% of the total peritoneal fluid cells) in (50–65% of the surface area covered; Fig. 2d). These macrophages the peritoneal fluid in the normal state, they were almost cleared were localized particularly on the surface of exposed fibrin clots, to occupy only 1% of peritoneal fluid cells by 4 h post-surgery consequently shielding ~60% of the surface area of exposed fibrin (Fig. 3a, b). This clearance was more evident at 24 h post-surgery, clots on day 3 (Supplementary Fig. 2c) and showed a phagocy- at which these cells were hardly detectable. We speculated that High − tosis ability (Supplementary Fig. 3a, b). This macrophage barrier this loss of F4/80 CD206 macrophages from the fluid was was temporal and cleared by day 5 when the mesothelial barrier due to the firm attachment of these cells to fibrin clots on the was reconstituted. ischemic button as described in the histological investigation + High − We found that F4/80 macrophages accumulating on the (Fig. 2a–d). The F4/80 CD206 macrophage subset was button surface were negative for CD206 (Supplementary Fig. 4). mostly reconstituted in the peritoneal fluid by day 5. On the other + Low + By contrast, CD206 cells, which were weakly positive for F4/80, hand, F4/80 CD206 macrophages were rarely found in the were observed within the ischemic button tissue after day 3, naive peritoneal fluid, but became detectable, occupying 3% of the indicating a distinction between these two macrophage subsets. peritoneal fluid cells, on day 1 with a peak at day 3. Throughout Low + Furthermore, at the lesion where an abdominal adhesion had the time course studied, the number of F4/80 CD206 mac- High − been formed between the ischemic button and surrounding tissue, rophages was similar to or smaller than that of F4/80 CD206 reduced accumulation of F4/80 macrophages on the fibrin clots macrophages (Fig. 3b). Both of these peritoneal macrophage was noted (Fig. 2e). These results collectively suggest that the cell subsets exhibited high expressions of tissue repair-related genes + − barrier composed of F4/80 CD206 macrophages on the and low expressions of pro-inflammatory or angiogenesis-related exposed fibrin clots may have the potential to contribute to genes (Supplementary Fig. 6a, b). As compared to the other High − attenuation of post-operative adhesion formation. subset, F4/80 CD206 macrophages were larger in size and exhibited higher mRNA or protein expression of Gata6, Tgfb2, Peritoneal macrophage subsets have distinct dynamics.We Ucp1, and ICAM2 (Fig. 3e and Supplementary Fig. 7a, b), which 30–32 then characterized the surface phenotype of peritoneal macro- are reported markers for resident peritoneal macrophages . Low + phages during the course of post-operative abdominal adhesion By contrast, the smaller F4/80 CD206 macrophages exhibited formation by using flow cytometry. Consistent with the above higher CCR2 and MHCII expression (Supplementary Fig. 7b), histological findings, two major macrophage subpopulations, F4/ suggesting their identity as monocyte-derived, recruited High − Low + 32,33 80 CD206 and F4/80 CD206 subsets, were identified macrophages . (Fig. 3a, b, and Supplementary Fig. 5). Both of these macrophage To examine the possibility that the other recruited monocytes/ Low − − + subsets, but not F4/80 CD206 , F4/80 CD206 or F4/80 macrophage subsets or granulocytes have the ability to cover the − − fibrin clot, we have further characterized the peritoneal cavity CD206 peritoneal cells, showed an evident phagocytosis 4 NATURE COMMUNICATIONS | (2021) 12:2232 | https://doi.org/10.1038/s41467-021-22536-y | www.nature.com/naturecommunications High - F4/80 CD206 Low + F4/80 CD206 Anti-F4/80-PE Anti-F4/80-PE Cell number (x10 ) Cell diameter (µm) NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-22536-y ARTICLE cells after ischemic button creation. The results showed that F4/ macrophages. This suggests that the fibrin clot itself has the Low − High 80 CD206 cells included Ly6C classical monocytes/ ability to attract resident peritoneal macrophages. Low macrophages, Ly6C non-classical monocytes/macrophages, We hypothesized that CD11b (integrin alpha M), which is + + 38 and Siglec-F eosinophils, while Ly6G neutrophils were found known to regulate the binding of microglia and fibrin , might − − in F4/80 CD206 population (Supplementary Fig. 8a, b). play a role in this interaction between resident peritoneal However, these cell types did not accumulate on the button macrophages and fibrin clots. A high-level CD11b expression of High − surface (Supplementary Fig. 9a). Furthermore, accumulated resident F4/80 CD206 peritoneal macrophages was con- F4/80 macrophages on fibrin clots were positive for a resident firmed (Supplementary Fig. 12b). Furthermore, proximity ligation macrophage marker, ICAM2 (Supplementary Fig. 9b, c). None- assay (PLA) detected clear PLA signals between fibrin and CD11b theless, there remained a concern that recruited monocytes/ on F4/80 macrophages (Fig. 5c), but not between fibrin and F4/ macrophages or granulocytes would not cover the fibrin clot 80 (Supplementary Fig. 12c). Furthermore, intraperitoneal because of their too-small occurrence at the time of initial injection of a blocking antibody to CD11b (clone; 5C6) adhesion formation (<24 h post-surgery). Therefore, we further intensified the abdominal adhesion, which corresponded to the assessed the ability of recruited monocytes/macrophages to abolished accumulation of F4/80 macrophages on the fibrin clot participate in the cell barrier formation in the “repeated ischemic (Fig. 5d–g). In addition, the above-mentioned transient disap- High − button creation model” (Supplementary Fig. 10a, b). In this pearance of resident F4/80 CD206 peritoneal macrophages model, we created an additional ischemic button on the right-side from the peritoneal fluid 1 day after surgery (Fig. 3a, b) was peritoneal wall on day 3 after the first ischemic button creation on inhibited by blocking CD11b, while the occurrence of recruited Low + the left-side peritoneal wall. This model enabled us to investigate F4/80 CD206 peritoneal macrophages in the peritoneal fluid the capability of recruited monocytes/macrophages more pre- was unaffected (Supplementary Fig. 12d, e). Thus, we conclude High − cisely because the occurrence of these recruited cells in the that accumulation of resident F4/80 CD206 peritoneal peritoneal cavity is much more abundant at day 3 compared to macrophages on the fibrin clot exposed on the damaged the time period for the ordinary cell barrier formation. The peritoneum was achieved directly via CD11b. obtained result demonstrated that, even with such an increased occurrence, recruited monocytes/macrophages or granulocytes did not contribute to the cell barrier formation. These results IL-4 reduces adhesions by enhancing macrophage barriers. together confirm that only resident macrophages have the ability Having identified the cell barrier function of resident F4/ High − to cover the fibrin clots. 80 CD206 peritoneal macrophages, this naturally-occurring process was not sufficient in preventing abdominal adhesions (Fig. 1 and Supplementary Fig. 1b, c). We thus explored if forced Depletion of resident macrophages exacerbates adhesions.We High − augmentation of resident F4/80 CD206 peritoneal macro- next investigated the functional role of resident peritoneal mac- phages would result in robust prevention of post-operative rophages in the post-operative adhesion formation by using a adhesion formation. We confirmed that intraperitoneal admin- 34,35 depletion strategy. Based on the previous reports , we adjusted istration of long-acting IL-4 complex (IL-4c) increased the the protocol of pre-injection of clodronate liposomes so that High − number of resident F4/80 CD206 peritoneal macrophages High − resident F4/80 CD206 peritoneal macrophages were depleted approximately three times (Fig. 6a, b). This IL-4c treatment at the time of surgery (Fig. 4a). This protocol did not reduce the indeed resulted in a markedly attenuated formation of post- Low + increase of recruited F4/80 CD206 peritoneal macrophages operative adhesions with increased coverage of the exposed fibrin post-surgery (Supplementary Fig. 11a, b). As a result of this clots by F4/80 macrophages (Fig. 6c–f). Histological assessments clodronate liposome treatment, we found markedly exacerbated demonstrated that mesothelial cells on the peritoneal membrane abdominal adhesions after ischemic button creation with a or fibrin deposition on the ischemic button lesion were unaffected reduced cover rate of fibrin clots by F4/80 macrophages by IL-4c injection (Fig. 6g–i). In addition, IL-4c administration (Fig. 4b–f). Of importance, the degree of the mesothelial dis- had no effect on other types of immune cells in the peritoneal ruption or the amount of fibrin deposition was unaffected cavity or within the ischemic button tissue (Supplementary (Fig. 4g–i and Supplementary Fig. 11c), suggesting that the Fig. 13a–c). Supporting this observation, IL-4c administration did High − depletion of resident F4/80 CD206 peritoneal macrophages not affect the Th1/Th2-related cytokine production levels in the did not affect fibrin formation or degradation. Consistent with peritoneal fluid 3 days after surgery, while a macrophage-derived this, the main pro- or anti-fibrinolytic factor, tissue plasminogen chemokine, CCL6/C10 , was up-regulated (Supplementary 36 37 activator (tPA) or plasminogen activator inhibitor-1 (PAI-1) , Fig. 14a, b). in the peritoneal fluid was not affected by the clodronate lipo- Consistent with the previous report that IL-4 promotes not some treatment (Fig. 4j, k). These data indicate that resident F4/ only resident macrophage proliferation but also recruited High − 80 CD206 peritoneal macrophages contribute to the macrophage polarization , IL-4c administration also increased attenuation of post-operative abdominal adhesion formation Low + the number of recruited F4/80 CD206 macrophages in the through reducing the exposure of fibrin clots as a cell barrier, but peritoneal fluid (Fig. 6a, b), while this cell number was less than not through affecting fibrin production or fibrinolysis. High − one-third of the F4/80 CD206 macrophage number. To Low + exclude the possibility that recruited F4/80 CD206 macro- Resident macrophages bind fibrin clots via CD11b. To under- phages contributed to the attenuation of post-operative adhe- stand the mechanism by which the fibrin clot attracted resident sions, we isolated each macrophage subsets from IL-4c injected peritoneal macrophages, we implanted an ex vivo produced fibrin mice and performed adoptive transfer into the peritoneal cavity of clot using a TISEEL fibrin glue kit (Baxter) into the mouse a syngeneic recipient mouse at the time of the ischemic button peritoneal cavity (Supplementary Fig. 12a). As a result, it was creation. As we expected, adoptive transfer of resident F4/ High − observed that a large number of resident peritoneal macrophages 80 CD206 peritoneal macrophages, but not recruited F4/ Low + accumulated onto the surface of the implanted exogenous fibrin 80 CD206 peritoneal macrophages, significantly reduced clot, similar to the finding of the fibrin clot on the ischemic post-operative adhesions (Fig. 6j, k). Accumulation of a sizeable High − button (Fig. 5a, b). A more than half of the exogenous fibrin clot number of transferred resident F4/80 CD206 peritoneal surface was covered by the accumulated resident peritoneal macrophages on the surface of the ischemic button was NATURE COMMUNICATIONS | (2021) 12:2232 | https://doi.org/10.1038/s41467-021-22536-y | www.nature.com/naturecommunications 5 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-22536-y b d PBS-Lipo Clod-Lipo PBS-Lipo Clod-Lipo -5 P = 5.23 x 10 *** DAPI 1.0 F4/80 0.8 Fibrin 0.6 -3 P = 5.05 x 10 0.4 ** 0.2 3.0 2.0 1.0 Zoom n.s. ef h j -7 n.s. P = 1.04 x 10 2.0 *** 100 1.5 DAPI PDPN F4/80 Fibrin 75 n.s. 1.0 50 0.5 Button Peritoneal cavity Day 0 Day 1 gi k n.s. n.s. 6.0 10.0 8.0 n.s. 4.0 6.0 50 n.s. 4.0 2.0 2.0 0 0 Day 0 Day 1 High − Fig. 4 Depletion of resident F4/80 CD206 peritoneal macrophages exacerbates post-operative abdominal adhesion formation. a The cell number High − of resident F4/80 CD206 peritoneal macrophages in the peritoneal cavity on day 2 after intraperitoneal injection of clodronate liposomes compared to the control PBS-liposome injection. n = 6 mice in each group. b, c The visual appearance of the abdominal adhesion formation (b) and the adhesion score (c) on day 1 post-ischemic button creation with an intraperitoneal injection of clodronate liposomes or PBS-liposomes two days before surgery. n = 15 mice in each group. d Representative images of the whole-mount immunohistostaining of the ischemic button on day 1 after surgery with pre-injection of clodronate liposomes or PBS-liposomes. n = 3 mice in each group. Scale bars, 100 μm. e–i Representative cross-section immunofluorescent images and quantification of the ischemic button on day 1 after surgery with the injection of clodronate liposomes or PBS-liposomes. The bar graphs show the coverage + + of exposed fibrin clots by F4/80 cells (f) or PDPN cells (g). The fibrin area (h) or thickness (i) on the ischemic button surface were also histologically quantified. n = 6 (PBS-Lipo group) and 7 (Clod-Lipo group) mice. Scale bars, 100 μm. j, k ELISA measurements for active tPA protein (j) or active PAI-1 protein (k) in the peritoneal fluid after surgery with pre-injection of clodronate liposomes or PBS-liposomes. n = 4 (Day 0) and 5 (Day 1) mice in each group. Data represent the mean ± SEM. **P < 0.01, ***P < 0.001, ns not significant, two-tailed student’s t-test. confirmed (Fig. 6l). These data proposed that the IL-4-based macrophages on their surface through CD11b-mediated binding. treatment would achieve substantial prevention of post-operative This macrophage accumulation results in the formation of a cell adhesion formation through augmentation of the resident barrier that acts to shield the adhesive fibrin clot from the sur- peritoneal macrophage barrier that shields exposed fibrin clots. rounding tissue in place of the lost mesothelial barrier, potentially attenuating the adhesion formation. The macrophage barrier is formed promptly by day 1 post-surgery and is temporal being replaced by the reconstructed mesothelial cell barrier by day 5. Discussion Compared to the complete mesothelial barrier, the macrophage Despite its enormous clinical importance, cellular and molecular barrier is insufficient; it covers only ~60% of the exposed fibrin processes underlying the formation and/or prevention of post- clot area. As a result, post-operative adhesions are developed at operative adhesions remains poorly understood . Here, in a the lesions that have an inadequate macrophage barrier. Of mouse model, we reveal a previously unidentified role of resident clinical impact, the resident macrophage barrier can be aug- peritoneal macrophages to attenuate post-operative adhesion mented by IL-4-based treatment or adoptive cell transfer, formation through the generation of a cell barrier (Fig. 7). Shortly which leads to robust prevention of the adhesion formation, after surgical insult, sticky fibrin clots are formed on the damaged proposing an innovative anti-adhesion strategy. In addition, peritoneal membrane, where the mesothelial barrier is lost. High − recruiting peritoneal macrophages are not involved in the These fibrin clots capture resident F4/80 CD206 peritoneal 6 NATURE COMMUNICATIONS | (2021) 12:2232 | https://doi.org/10.1038/s41467-021-22536-y | www.nature.com/naturecommunications PBS-Lipo Clod-Lipo PBS-Lipo Clod-Lipo PBS-Lipo Clod-Lipo PBS-Lipo PBS-Lipo Clod-Lipo Clod-Lipo PBS-Lipo Clod-Lipo Clod-Lipo PBS-Lipo High - F4/80 CD206 cell number (x10 ) Adhesion score Fibrin cover by Fibrin cover by PDPN cells (%) + F4/80 cells (%) Fibrin thickness Fibrin area on (µm) button surface (%) Active PAI-1 (ng/ml) Active tPA (ng/ml) PBS-Lipo PBS-Lipo Clod-Lipo Clod-Lipo PBS-Lipo PBS-Lipo Clod-Lipo Clod-Lipo NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-22536-y ARTICLE a Surface of b Cross-section of exogenous fibrin clot Zoom exogenous fibrin clot Exogenous fibrin clot DAPI DAPI F4/80 F4/80 Fibrin Fibrin Peritoneal cavity -2 P = 3.14 x 10 cd Anti-CD11b e IgG2b PLA 3.0 Fibrin CD11b (Fibrin / CD11b) 2.0 Button 1.0 Peritoneal cavity DAPI F4/80 Fibrin g PLA Fibrin -2 F4/80 Zoom P = 2.47 x 10 Button CD11b F4/80 IgG2b Peritoneal cavity Anti- CD11b High − Fig. 5 Resident F4/80 CD206 peritoneal macrophages bind fibrin clots via CD11b. a, b Representative confocal image of the whole-mount (a) and cross-section (b) immunofluorescent staining of the implanted exogenous fibrin clot on day 1 after implantation. n = 3 mice. Scale bars, 100 μm. (c) Representative confocal images of PLA between CD11b of F4/80 macrophages and fibrin on the ischemic button surface. PLA signals are shown in red. n = 3 independent experiments. Scale bars, 5 μm. d, e Representative images of abdominal adhesions (d) and adhesion scores (e) on day 1 post-ischemic button creation with an intraperitoneal injection of control IgG2b antibody (IgG2b group, n = 6 mice) or anti-CD11b blocking antibody (Anti-CD11b group, n = 7 mice). f, g Representative cross-section immunohistostaining images of the ischemic button (f) and quantification of the coverage of exposed fibrin clots by F4/80 cells (g) on day 1 after surgery with an intraperitoneal injection of control IgG2b antibody or anti-CD11b blocking antibody. n = 5 mice in each group. Scale bar, 100 μm. Data represent the mean ± SEM. *P < 0.05, two-tailed Student’s t-test. protective cell barrier mechanism, shedding light on a previously disappearance reaction (MDR) in the setting of intra-abdominal 42,43 unknown biological insight regarding the complex diversity of infection . It is also known that intraperitoneal injection of 30 35 44 macrophages. lipopolysaccharide , zymosan , or thioglycolate induces the We here demonstrate that the accumulation of resident peri- MDR and that the MDR is suppressed by pretreatment with 35,42 toneal macrophages on the injured peritoneum is mediated anticoagulants, including warfarin and heparin . The under- through binding between CD11b and fibrin. There is a report pinning mechanism of the MDR is thought to be macrophage describing that peritoneal macrophages are attracted to the accumulation to a foreign body or migration into the lymphatic 35,43 damaged tissue through sensing extracellular ATP and contribute vessels, rather than the death of macrophages . Our study to the removal of dead cells in a heat-induced liver injury model reports that the MDR occurs during post-operative adhesion 34 High − in mice . However, we observe that implantation of an formation as well. Resident F4/80 CD206 peritoneal mac- externally-produced fibrin clot, which is unlikely to release a rophages that occupy around 40% of the total peritoneal fluid significant level of ATP, results in macrophage accumulation on cells in the normal status are almost cleared as early as day 1 after its surface, similar to that observed in the ischemic button model. surgery. The series of histological examinations in this study Moreover, in addition to the positive PLA signals between CD11b suggest that this clearance from the peritoneal fluid is due to and fibrin, the inhibition of CD11b-fibrin interaction using a the attachment of these cells to the fibrin clots on the ischemic blocking antibody abolishes the accumulation of resident mac- button. By day 5 after surgery, the population of resident High − rophages onto the damaged peritoneum of the ischemic button. F4/80 CD206 peritoneal macrophages is reconstituted in the Taken together, although a possible role of extracellular ATP and peritoneal fluid. Although the source of these reconstituting other factors cannot be fully excluded, we conclude that the macrophages remains uncertain, it could be their detachment CD11b-fibrin binding is a major mechanism by which resident from the fibrin clots, migration from the peritoneal organs , peritoneal macrophages accumulated onto the injured perito- differentiation from recruited monocytes, and/or phenotype Low + neum in our model of post-operative adhesion. change from recruited F4/80 CD206 peritoneal macrophages 46–48 One of the interesting findings in this study is that resident or other cells types . peritoneal macrophages drastically but transiently disappeared Peritoneal macrophages are classified into two major distinct from the peritoneal fluid after adhesion-inducing surgery. A subsets, the large peritoneal macrophage (LPM) and small peri- similar phenomenon has been reported as the macrophage toneal macrophage (SPM), basically according to their size . NATURE COMMUNICATIONS | (2021) 12:2232 | https://doi.org/10.1038/s41467-021-22536-y | www.nature.com/naturecommunications 7 IgG2b Anti-CD11b IgG2b Anti-CD11b Fibrin cover by Adhesion score F4/80 cells (%) ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-22536-y a b PBS IL-4c -2 P = 1.91 x 10 10 2.0 2.0 -3 P = 2.18 x 10 1.5 3 1.5 ** 1.0 1.0 0.5 0.5 3 4 3 4 010 10 010 10 0 PBS IL-4c PBS IL-4c Anti-CD206-Alexa488 c ef h -2 P = 2.43 x 10 PBS IL-4c PBS IL-4c n.s. Button 100 DAPI 75 F4/80 50 50 25 25 Peritoneal cavity 0 0 DAPI PBS IL-4c -2 PBS IL-4c P = 2.34 x 10 PDPN g i n.s. F4/80 3.0 Fibrin 100 15 2.0 n.s. 1.0 50 Zoom 25 0 0 PBS IL-4c PBS IL-4c PBS IL-4c j l High - Low + PBS F4/80 CD206 F4/80 CD206 Day 1 Day 3 Day 5 Button DAPI F4/80 CM-DiI Fibrin Peritoneal cavity n.s. -2 P = 3.32 x 10 3.0 Zoom 2.0 1.0 Fig. 6 IL-4c administration augments the macrophage barrier and attenuates post-operative adhesion formation. a, b Representative flow cytometry plots (a) and cell numbers (b) of peritoneal macrophage subsets day 2 after injection of IL-4c or PBS only. n = 5 mice in each group. c, d The representative appearance of post-operative adhesions (c) and the adhesion score (d) on day 3 post-surgery. IL-4c or PBS was administered into the peritoneal cavity immediately after ischemic button creation surgery. n = 9 mice per group. e–i Representative immunofluorescence images and quantification of cross- sections of the ischemic button on day 3 after surgery with IL-4c or PBS administration. The bar graphs show the coverage of exposed fibrin clots by F4/80 + + cells (f) or PDPN cells (g). The fibrin area (h) or thickness (i) on the ischemic button surface were also histologically measured. n = 5 (PBS) and 7 (IL- 4c) mice. Scale bars, 100 μm. j, k Representative images of post-operative abdominal adhesions (j) and calculated adhesion scores (k) on day 5 after High − Low + ischemic button creation with intraperitoneal adoptive transfer of syngeneic mouse F4/80 CD206 or F4/80 CD206 peritoneal macrophages, or High − Low + PBS administration. n = 5 (PBS), 4 (F4/80 CD206 ) and 5 (F4/80 CD206 ) mice. l Representative immunofluorescence images of the ischemic High − button with adoptive transfer of resident F4/80 CD206 peritoneal macrophages (pre-labeled with CM-DiI; red). n = 2 independent experiments. Scale bar, 100 μm. Data represent the mean ± SEM. *P < 0.05, **P < 0.01, ns not significant, one-way ANOVA and Tukey’s post hoc test in (k) or two-tailed student’s t-test in others. Low Low High Low LPMs are the resident macrophages, which are dominant at the exhibit an F4/80 , CD11b , MHCII , GATA6 , and steady-state and maintained in the peritoneal cavity through self- CD206 phenotype. Sexually dimorphic replenishment of LPM High High Low renewal, and have an F4/80 , CD11b , MHCII , by bone marrow-derived cells, which is high in males and very High − 46,49 48 GATA6 , and CD206 phenotype . By contrast, SPMs are low in females, has also been reported . In addition, there is the recruited macrophages, which are derived from circulating information available regarding the functional diversity of these monocytes and increase in number in response to an inflam- subsets. For instance, LPMs regulate the peritoneal B-1 cell- matory stimulus in a CCR2 dependent manner . These cells mediated gut IgA production through TGFβ2 under the normal 8 NATURE COMMUNICATIONS | (2021) 12:2232 | https://doi.org/10.1038/s41467-021-22536-y | www.nature.com/naturecommunications PBS High - F4/80 CD206 Low + F4/80 CD206 Anti-F4/80-PE Adhesion score Adhesion score High - F4/80 CD206 cell number (x10 ) Low + F4/80 CD206 Fibrin cover by Fibrin cover by Cell number (x10 ) + + PDPN cells (%) F4/80 cells (%) Fibrin area on Fibrin thickness the button surface (%) (µm) NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-22536-y ARTICLE Surgical trauma Insufficient macrophage barrier Mesothelial loss & Fibrin clot still exposed Adhesion fibrin clot formation formation Sufficient macrophage barrier Mesothelial cells High - F4/80 CD206 macrophages + IL-4c Fibrin clot Fibrin clot covered by macrophages Re-mesothelialization Fig. 7 The function of macrophage barriers during adhesion formation. The schematic illustrates the role of resident macrophage barriers during post- High − operative abdominal adhesion formation. Upon injury of the peritoneal membrane, resident F4/80 CD206 peritoneal macrophages promptly accumulate onto the lesion where fibrin clots are exposed. These accumulated macrophages form an anti-adhesion cell barrier to shield the sticky fibrin clots from neighboring tissues. This macrophage barrier is, however, usually insufficient, allowing the formation of adhesions (Upper panels). However, when this cell barrier is sufficiently strengthened i.e., using IL-4c treatment, post-operative adhesions are robustly prevented (Lower panels). condition . SPMs have an ability to present antigens to naive responses to IL-4 and the resulting effects on adhesion formation + 50 CD4 T cells , suggesting that SPMs induce local T-cell priming could be elucidated by using cell-type-specific IL-4 receptor during peritoneal infection or inflammation. In our study, knockout mice. Of note, a previous study has reported that IL-4 High − Low + 24 F4/80 CD206 and F4/80 CD206 macrophage subsets knockout did not affect post-operative adhesions in mice , exhibited comparable characters to LPMs and SPMs, respectively, indicating that endogenous IL-4 is unlikely to play a critical role 30,33,51 in terms of morphology and gene/protein expression . in the adhesion formation. However, further investigation is needed to define these macro- The findings obtained in this study have significant clinical phages as LPMs and SPMs. implications in developing an innovative anti-adhesion strategy. + + We observed that sizeable numbers of CD11b CCR2 To prevent post-operative adhesion formation, a range of phar- High + + Ly6C classical monocytes/macrophages and CD11b CCR2 macological agents, including fibrinolytic agents, non-steroidal Low Ly6C non-classical monocytes/macrophages, which were F4/ anti-inflammatory drugs, steroids, antioxidants, histamine Low − 80 CD206 , appeared in the peritoneal cavity at day 3 post- antagonists, and statins, were tested; however, none of them has High − Low + 53 surgery. Unlike F4/80 CD206 and F4/80 CD206 mac- been clinically successful to date . Instead, implantation/ Low − rophage subsets, these F4/80 CD206 cell populations did not administration of a biomaterial barrier product is currently present a phagocytic ability, suggesting that they were unlikely to conducted for the prevention of adhesions . Nonetheless, this be mature macrophages. Furthermore, our results showed that method has obvious limitations in its efficacy and + 8–10 CCR2 recruited monocytes/macrophages did not have the practicability . Macrophage barriers are likely to have multiple ability to form an anti-adhesion cell barrier. On the other hand, important advantages over the biomaterial-based barrier. The there is a report showing that pre-treatment of rabbits with formation of a macrophage barrier relies on a natural, physiolo- proteose peptone or transplantation of proteose peptone-elicited gical healing process, while an artificial biomaterial barrier is peritoneal macrophages protect against adhesion formation, associated with possible foreign body reactions . The macro- suggesting that monocyte-derived inflammatory macrophages phage barrier has no risk of physical displacement, unlike a 21 9,10 might offer an anti-adhesion effect in a certain condition .Itis biomaterial barrier . In addition, the mobility and flexibility of speculated that this effect of monocyte-derived macrophages may macrophages enable the formation of a cell barrier at any location be attained through a different mechanism(s) from cell barrier throughout the peritoneal cavity, including difficult-to-access formation, including modulation of inflammation, fibrin forma- localizations, i.e., deep inside of the pelvic cavity or between tion, fibrinolysis and/or vacuolization, and fibrosis of adhesion intestines. Furthermore, sufficient macrophage barriers can be 21–25 tissues . achieved simply by injection of a pharmacological reagent, i.e., IL- IL-4 is known to play pleiotropic roles in the immune system . 4c, and is thereby accessible in any type of surgical procedures, We found that intraperitoneal IL-4c administration attenuated including open, laparoscopic, or robotic surgeries. Further post-operative adhesion formation by increasing the number of development of this innovative treatment to prevent post- High − F4/80 CD206 resident peritoneal macrophages. Adoptive operative adhesions, which is based on the previously unknown High − transfer of IL-4c-treated F4/80 CD206 macrophages resulted concept of the macrophage barrier, is warranted. To this end, it is in comparable attenuation of adhesions. On the other hand, IL-4c essential to confirm the universality of the results obtained in this administration did not affect the number of non-macrophage cell study. We here used the mouse ischemic button model because it types including lymphocytes or the major Th1/Th2-related has an important advantage in inducing reproducible intra- cytokine levels in the peritoneal fluid. Therefore, we consider that abdominal adhesions compared to other models as previously the resident peritoneal macrophage barrier is the central demonstrated . Indeed, this model offered reliable and adequate mechanism for the anti-adhesion effect of IL-4c administration. adhesion formation to our study. As the next step toward clinical However, we cannot exclude a possibility of functional alterations application, investigations using a different model, particularly a of these cells in response to IL-4, which would modulate a process large animal model, and also research with human cells will be of adhesion formation. Further detailed cell-type dependent required. NATURE COMMUNICATIONS | (2021) 12:2232 | https://doi.org/10.1038/s41467-021-22536-y | www.nature.com/naturecommunications 9 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-22536-y Methods Peritoneal cell isolation and flow cytometry analysis. After mice were sacrificed Study approval. All investigations using living animals conformed to the Princi- using CO inhalation, peritoneal cells were obtained by washing the peritoneal cavity with ice-cold PBS four times. Erythrocytes were removed by incubating with ples of Laboratory Animal Care formulated by the National Society for Medical Research and to NIH guidelines (Guide for the Care and Use of Laboratory Ani- red blood cell lysis buffer (Biolegend, 420301) at 5 min on ice. Total live cell number was counted using a Countess II Automated Cell Counter (Invitrogen, mals. National Academies Press 1996) and performed with the approval of the ethics committee of the Queen Mary University of London and the UK Home AMQAX1000) with Trypan Blue staining. After centrifugation, peritoneal cells were Office (Project License PPL70/8503). All in vivo procedures were performed by UK resuspended in FACS buffer (HBSS plus 2 mM EDTA and 0.5% BSA) and pre- Home Office Personal License holders. In vivo and in vitro procedures and incubated with an anti-mouse CD16/CD32 antibody (clone: 93) (1:100 dilution, assessments were blinded wherever possible. Invitrogen, 14–0161–85) to block the Fc receptor. Dead cells and debris were excluded by forwarding scatter/side scatter (FSC/SSC) and DAPI staining (1:500 dilution). Phenotype analysis was performed by staining peritoneal cells with the fluorophore-conjugated antibodies for 3 h at 4 °C. Information about primary Post-operative abdominal adhesion model. Ten- to twelve-week-old male antibodies and dilutions used in the assay can be found in “Supplementary Table 1”. C57BL/6 mice were purchased from Charles River Laboratories UK and used in the In order to accurately identify the positive signal, appropriate isotype antibodies experiments. These mice were maintained in a specific pathogen-free room under a were used as a negative control. Flow cytometric analyses were performed using a 12-h light/12-dark cycle, 20–24 °C temperature, and 45–55% humidity with free BD LSRFortessa cell analyzer (BD Biosciences) with FlowJo software version 10 access to food and water in our animal facility. Surgery was performed under (Tree Star). Exemplifying the gating strategy is provided in the Supplementary aseptic conditions, and all efforts were made to minimize animal suffering. Information (Supplementary Fig. 5). Cell sorting was performed using a FACS Aria Anesthesia was induced by 2.0% isoflurane inhalation. Abdominal adhesion for- High − II (BD Biosciences). To measure the cell size of individual F4/80 CD206 and mation was induced by the construction of two ischemic buttons on the peritoneal Low + F4/80 CD206 macrophages, the digital images of sorted cells were acquired with wall, as previously described . After a small median laparotomy, two ischemic an All-in-One microscope (BZ-8000; KEYENCE) using a 20× objective lens (1360 × buttons, ~5 mm in diameter each and spaced ~10 mm apart, were created on the 1024 pixels). Cell diameter was determined using ImageJ software. right-side peritoneal wall by clamping and ligating with a 5–0 Mersilk suture (W2500T, Ethicon). The abdomen was closed with a double-layered suture of the peritoneum and skin with a 5–0 vicryl absorbable suture (W9915, Ethicon). They Whole-mount staining. Tissues or exogenous fibrin clots were fixed in 4% PFA were allowed to recover on a heating pad and monitored daily until euthanasia. At overnight at 4 °C and permeabilized and blocked in PBS containing 0.5% Triton X- the chosen time point, the mice were sacrificed with carbon dioxide (CO ) inha- 100 and 5% BSA for 4 h at 4 °C. Subsequently, samples were labeled with primary lation to evaluate abdominal adhesion formation, and peritoneal fluid/tissue antibodies overnight at 4 °C. After rinsing in PBS three times for 5 min, the sections 4,28 samples collected. According to the adhesion scoring system , the adhesion were next incubated with DAPI and the appropriate fluorophore-conjugated sec- scores for each ischemic button (Lower; tail side, Upper; head side) were deter- ondary antibodies overnight at 4 °C. Confocal images were captured in z-stacks of mined as follows: 0, no adhesion; 1, filmy adhesion (separated easily by blunt 9–13 planes and reconstructed with the maximum intensity projection using ZEN dissection); 2, firm adhesion (separated by aggressive blunt dissection); 3, dense black 3.0 lite software (Carl Zeiss). All antibodies using whole-mount staining are adhesion (separable only by sharp dissection). Adhesion scores were validated in a described in Supplementary Table 1. blinded manner, and the average of lower and upper ischemic button scores was calculated (Supplementary Fig. 1a). For the repeated ischemic button creation Administration of liposomes, blocking antibodies and IL-4c. For macrophage model (Supplementary Fig. 10a), two ischemic buttons were first created on the depletion, 50 μg of clodronate liposomes (Clod-Lipo) or control liposomes (PBS- left-side peritoneal wall of a mouse as described above. Three days after this first Lipo) (Encapsula NanoSciences, CLD-8901) were injected intraperitoneally 2 days surgery, the abdomen was re-opened, and an additional ischemic button was before surgery. In order to determine the phagocytosis ability of peritoneal cavity created on the right-side peritoneal wall. cells, 50 μg of Fluoroliposome-DiD (Encapsula Nano Sciences, CLD-8913) were injected intraperitoneally 2 days after surgery. For blocking macrophage-fibrin interaction, 50 μg anti-CD11b neutralizing antibody (clone: 5C6) (Invitrogen, MA5- Hematoxylin and eosin staining and immunohistolabeling. After harvesting, 16528) or rat IgG2b control antibody (clone: eB149/10H5; eBioscience,16–4031–85) tissues were fixed in 4% paraformaldehyde (PFA) overnight at 4 °C and were was injected intraperitoneally 3 h before and immediately after surgery. For embedded and frozen in O.C.T. compound (VWR International). increasing macrophages, IL-4c (5 μg recombinant IL-4 (PeproTech, 214–14) and 25 For H&E staining of ischemic buttons and adhesion tissues, 6 µm thick frozen μg stabilizing monoclonal anti-IL-4 antibody (clone: BVD4-1D11; BD Biosciences, sections were cut and incubated in Mayer’s hematoxylin solution (Sigma-Aldrich, 554387)) dissolved in 100 μl PBS were intraperitoneally injected to normal mice or MHS16) for 15 min and Eosin Y solution (Sigma-Aldrich, HT110116) for 1 min. administrated into the peritoneal cavity immediately after the ischemic button After dehydration through increasing concentrations of ethanol to xylene, the creation. An equivalent volume of PBS was injected as a control. sections were mounted using the DPX mounting medium (VWR International, 13512). The digital images were acquired with an All-in-One microscope (BZ-8000; KEYENCE) using a 2× or 20× objective lens (1360 × 1024 pixels). Active tPA and PAI-1 measurements. Peritoneal fluid from the adhesion model For immunohistolabeling, frozen sections were cut at 6 µm thick and pre- mice that were pre-injected with clodronate liposomes or control liposomes was blocked with the blocking buffer (PBS plus 5% BSA and 0.1% Tween 20) for 30 collected through washing the peritoneal cavity with 500 μl ice-cold PBS. Protein min at room temperature. To block endogenous biotin, if required, samples were concentrations of the peritoneal fluid were determined using a DC Protein Assay pre-blocked with streptavidin buffer (PBS plus 0.1 mg/ml Streptavidin and 0.1% (Bio-Rad, 5000112). ELISA was performed according to the manufacturer’s Tween 20) for 30 min and biotin buffer (PBS plus 0.5 mg/ml Biotin and 0.1% instructions of mouse active tPA functional assay ELISA kit (Innovative Research, Tween-20) for 30 min at room temperature. Samples were labeled with the IMSTPAKTA) and active mouse PAI-1 functional assay ELISA kit (Innovative primary antibodies overnight at 4 °C. Information about primary antibodies and Research, IMSPAI1KTA). dilutions used in the assay can be found in “Supplementary Table 1”.The anti- laminin antibody (1:1000 dilution, Sigma, L9393) was conjugated with Alexa Exogenous fibrin clot implantation into the mouse abdomen.Exogenous fibrin Fluor 488 fluorophores using labeling kits (Life Technologies, A20181) clots for implantation were produced using TISSEEL (Baxter, 1502243). Briefly, 10 μl according to the manufacturer’s instruction. After rinsing in PBS three times for of component 1 (Human fibrinogen, 72–110 mg/ml; Aprotinin (synthetic), 3000 5 min, the sections were next incubated with 4′,6-diamidino-2-phenylindole KIU/ml) and 10 μl of component 2 (Human thrombin, 500 IU/ml; Calcium chloride (DAPI) (Sigma-Aldrich, D9542) and the appropriate fluorophore-conjugated dihydrate, 40 µmol/ml) were mixed on a sliding glass using a pipette tip. For easy secondary antibodies or AlexaFluor 546-conjugated Streptavidin (1:300 dilution, identification of exogenous fibrin clots in histological analyses, 10 mg/ml Alexa Fluor Invitrogen, S11225) for 1 h at the room temperature. Stained sections were 647-conjugated human fibrinogen (Invitrogen, F35200) was mixed in component 1 mounted with DAKO Fluorescence Mounting Medium (Agilent, S302380-2), (1:500 dilution). After fibrin clot formation, the exogenous clots (~5 mm in dia- and the digital images were acquired, and pseudo-colored with an All-in-One meter) were washed five times with PBS to remove excess salt and implanted into the microscope (BZ-8000; KEYENCE) using a 10× or 20× objective lens (1360 × peritoneal cavity of a mouse through a laparotomy (Supplementary Fig. 12a). 1024 pixels). All image analysis was performed by importing images as TIFF files into ImageJ version 1.50i software (NIH). To visualize a whole image of the ischemic button Proximity ligation assay. Interaction between CD11b and fibrin was examined (Fig. 2a), sixteen images were reconstructed by using the MosaicJ plugin. For the using reagents from the Duolink proximity ligation assay (Sigma-Aldrich, + + quantification of the cover ratio by F4/80 macrophages or PDPN mesothelial DUO92106) following the manufacturer’s instructions. Briefly, 4% PFA-fixed cells on the ischemic button and fibrin clots, F4/80 or PDPN positive area was frozen sections were cut at 6 µm thick and pre-blocked with the blocking buffer quantified as the percentage of the surface of peritoneal area or fibrin positive area. (PBS plus 5% BSA and 0.1% Tween 20) for 30 min at room temperature. Samples For quantification of fibrin formation, the fibrin area and thickness were quantified were labeled with anti-CD11b antibody (clone: M1/70; 1:200 dilution, eBioscience, as the percentage or average thickness of the fibrin positive area on the ischemic 14–0112–85) and anti-fibrin antibody (1:500 dilution, Abcam, ab34269) overnight button (Supplementary Fig. 11c). Images containing the surface area of the at 4 °C. After rinsing in PBS three times for 5 min, the sections were incubated with ischemic button, excluded adhesion area, were acquired at three independent Goat AlexaFluor 488-conjugated anti-rat IgG antibody (1:300 dilution, Invitrogen, regions and analyzed. A color threshold function of ImageJ was applied to measure A11006) for 1 h at room temperature. Slides were washed and incubated with a macrophages, mesothelial cells, and fibrin-clot area. PLA probe antibody, which is conjugated with PLUS or MINUS short nucleotide 10 NATURE COMMUNICATIONS | (2021) 12:2232 | https://doi.org/10.1038/s41467-021-22536-y | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-22536-y ARTICLE sequences, for 1 h at 37 °C. PLUS and MINUS oligonucleotides were ligated and 2. Menzies, D. & Ellis, H. Intestinal obstruction from adhesions-how big is the amplified by rolling circle PCR with fluorescent nucleotides for 90 min at 37 °C. problem? Ann. R. Coll. Surg. Engl. 72,60–63 (1990). After rolling circle amplification, slides were washed and incubated with APC-anti- 3. Stommel, M. W. J. et al. Multicenter observational study of adhesion F4/80 antibody (clone: BM8; 1:200 dilution, Biolegend, 123116) and Goat Alexa- formation after open-and laparoscopic surgery for colorectal cancer. Ann. Fluor 405-conjugated anti-rabbit IgG antibody (1:300 dilution, Invitrogen, A31556) Surg. 267, 743–748 (2018). for 1 h at room temperature. F4/80 and fibrin interaction was examined as a 4. Ten Broek, R. P. G. et al. Bologna guidelines for diagnosis and management of negative control, in the same manner, using anti-F4/80 antibody (clone: BM8; adhesive small bowel obstruction (ASBO): 2017 update of the evidence-based 1:200 dilution, eBioscience, 14–4801–81) and APC-anti-CD11b antibody (clone: guidelines from the world society of emergency surgery ASBO working group. M1/70; 1:200 dilution, eBioscience, 17–0112–82). PLA was visualized using an World J. Emerg. Surg. 13, 24 (2018). inverted Zeiss 800 (Carl Zeiss) confocal laser scanning microscope equipped with 5. Sutton, C. Adhesions and their prevention. Obstet. Gynaecol. 7, 168–176 63× (1.4 NA) objective and solid-state laser diodes (405, 488, 561, and 640 nm (2005). excitation wavelengths). 6. Wiseman, D. M., Trout, J. R. & Diamond, M. P. The rates of adhesion development and the effects of crystalloid solutions on adhesion development Adopted macrophage transfer. Peritoneal cavity cells were collected from four IL- in pelvic surgery. Fertil. Steril. 70, 702–711 (1998). 7. Stapleton, L. M. et al. Use of a supramolecular polymeric hydrogel as an 4c administrated, syngeneic donor mice on day 5 post-ischemic button creation. High − Low + F4/80 CD206 and F4/80 CD206 macrophages were isolated using a FACS effective post-operative pericardial adhesion barrier. Nat. Biomed. Eng. 3, Aria II (BD Biosciences), as described above. We transferred 400,000 cells in 100 μl 611–620 (2019). PBS per recipient mouse directly into the peritoneal cavity immediately after sur- 8. Ten Broek, R. P. G. et al. Benefits and harms of adhesion barriers for gery. After 5 days, the mice were sacrificed and adhesion scores evaluated. For abdominal surgery: a systematic review and meta-analysis. Lancet 383,48–59 High − enabling histological tracing of transferred macrophages, isolated F4/80 CD206 (2014). peritoneal macrophages were labeled with CM-DiI before transplantation (Invi- 9. Brochhausen, C. et al. Current strategies and future perspectives for trogen, C7000) according to the manufacturer’s recommendations. intraperitoneal adhesion prevention. J. Gastrointest. Surg. 16, 1256–1274 (2012). 10. Trochsler, M. & Maddern, G. J. Adhesion barriers for abdominal surgery: a sticky problem. Lancet 383,8–10 (2014). RNA extraction and real-time PCR. Total RNA was extracted using the PicoPure™ High − 11. Koninckx, P. R., Gomel, V., Ussia, A. & Adamyan, L. Role of the peritoneal RNA Isolation Kit (Applied Biosystems, KIT0204) from F4/80 CD206 and Low + cavity in the prevention of postoperative adhesions, pain, and fatigue. Fertil. F4/80 CD206 macrophages, which were isolated from the peritoneal cavity Steril. 106, 998–1010 (2016). 5 days post-surgery using a cell sorter as described above. The RNA concentration 12. Mutsaers, S. E., Prele, C. M., Pengelly, S. & Herrick, S. E. Mesothelial cells and and quality were measured using the DS-11 Spectrophotometer (DeNovix). cDNA peritoneal homeostasis. Fertil. Steril. 106, 1018–1024 (2016). was prepared using the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, 4368814) according to the manufacturer’s instruction. Real-time PCR 13. diZerega, G. S. & Campeau, J. D. Peritoneal repair and post-surgical adhesion was performed by Quant studio 7 qPCR machine (Applied Biosystems) with a formation. Hum. Reprod. Update 7, 547–555 (2001). PowerUP SYBR Green Master Mix (Applied Biosystems, A25742). Bone marrow 14. Maciver, A. H., McCall, M. & James Shapiro, A. M. Intra-abdominal mononuclear cells (BMMNCs) were collected from the femurs and tibiae of normal adhesions: cellular mechanisms and strategies for prevention. Int J. Surg. 9, mice as previously described and were used as a control. Fold changes were 589–594 (2011). calculated by relative quantification based on the second derivative method. Gene 15. Hellebrekers, B. W. & Kooistra, T. Pathogenesis of postoperative adhesion expression levels were normalized by the Gapdh gene. All primers are described in formation. Br. J. Surg. 98, 1503–1516 (2011). “Supplementary Table 2”. 16. Kosaka, H., Yoshimoto, T., Yoshimoto, T., Fujimoto, J. & Nakanishi, K. Interferon-gamma is a therapeutic target molecule for prevention of postoperative adhesion formation. Nat. Med 14, 437–441 (2008). Cytokine profiling. The peritoneal fluid was collected through washing the peri- 17. Braun, K. M. & Diamond, M. P. The biology of adhesion formation in the toneal cavity with 500 μl ice-cold PBS of the mice on day 3 post-ischemic button peritoneal cavity. Semin Pediatr. Surg. 23, 336–343 (2014). creation surgery with IL-4c or PBS administration. After determining the protein 18. Tsai, J. M. et al. Surgical adhesions in mice are derived from mesothelial cells concentration using a DC Protein Assay (Bio-Rad, 5000112), the peritoneal fluid and can be targeted by antibodies against mesothelial markers. Sci. Transl. containing 20 μg total protein was pooled from 6 mice (total 120 μg protein for Med 10, eaan6735 (2018). each group). Cytokine profiling was performed according to the manufacturer’s instructions of Mouse XL Cytokine Array Kit (R&D Systems, ARY028). Briefly, 19. Tsai, J. M. et al. Neutrophil and monocyte kinetics play critical roles in mouse peritoneal adhesion formation. Blood Adv. 3, 2713–2721 (2019). following the block with blocking buffer at room temperature for 1 h, the mem- branes were incubated with pooled peritoneal fluids overnight at 4 °C. After rinsing 20. Fischer, A. et al. Post-surgical adhesions are triggered by calcium-dependent with washing buffer three times for 10 min, membranes were incubated with a membrane bridges between mesothelial surfaces. Nat. Commun. 11, 3068 biotinylated detection antibody cocktail at room temperature for 1 h and then (2020). incubated with streptavidin–horseradish peroxidase for 30 min. Membranes were 21. Ar’Rajab, A. et al. Enhancement of peritoneal macrophages reduces exposed to the peroxidase substrate and measured with the Azure c600 Imager postoperative peritoneal adhesion formation. J. Surg. Res 58, 307–312 (1995). (Azure Biosystems). The images were analyzed using ImageJ software. 22. Burnett, S. H. et al. Development of peritoneal adhesions in macrophage depleted mice. J. Surg. Res 131, 296–301 (2006). 23. Hoshino, A. et al. Inhibition of CCL1-CCR8 interaction prevents aggregation Statistics. All statistical tests were performed using RStudio version 1.0.153 soft- of macrophages and development of peritoneal adhesions. J. Immunol. 178, ware and GraphPad Prism 8 software (GraphPad Software). Data represent the 5296–5304 (2007). mean ± SEM. For comparisons between multiple groups, one-way ANOVA was 24. Hong, G. S. et al. Effects of macrophage-dependent peroxisome proliferator- performed, followed by Tukey’s post hoc test. Two groups were compared using activated receptor gamma signalling on adhesion formation after abdominal the two-tailed, unpaired Student’s t-test. surgery in an experimental model. Br. J. Surg. 102, 1506–1516 (2015). 25. Honjo, K. et al. Plasminogen activator inhibitor-1 regulates macrophage- Reporting summary. Further information on research design is available in the Nature dependent postoperative adhesion by enhancing EGF-HER1 signaling in mice. Research Reporting Summary linked to this article. FASEB J. 31, 2625–2637 (2017). 26. Beyene, R. T., Kavalukas, S. L. & Barbul, A. Intra-abdominal adhesions: Anatomy, physiology, pathophysiology, and treatment. Curr. Probl. Surg. 52, Data availability 271–319 (2015). Data that support the findings of this study are available within the article and Supplementary Information or from the corresponding author upon reasonable request. 27. Buckenmaier, C. C. 3rd, Pusateri, A. E., Harris, R. A. & Hetz, S. P. Comparison Raw source data are present in the Source Data file Source data are provided with of antiadhesive treatments using an objective rat model. Am. Surg. 65, 274–282 (1999). this paper. 28. Coccolini, F. et al. Peritoneal adhesion index (PAI): proposal of a score for the “ignored iceberg” of medicine and surgery. World J. Emerg. Surg. 8, 6 (2013). Received: 5 August 2020; Accepted: 15 March 2021; 29. Lua, I., Li, Y., Pappoe, L. S. & Asahina, K. Myofibroblastic conversion and regeneration of mesothelial cells in peritoneal and liver fibrosis. Am. J. Pathol. 185, 3258–3273 (2015). 30. Okabe, Y. & Medzhitov, R. Tissue-specific signals control reversible program of localization and functional polarization of macrophages. Cell 157, 832–844 References (2014). 1. Krielen, P. et al. Adhesion-related readmissions after open and laparoscopic 31. Gautier, E. L. et al. Gene-expression profiles and transcriptional regulatory surgery: a retrospective cohort study (SCAR update). Lancet 395,33–41 pathways that underlie the identity and diversity of mouse tissue (2020). macrophages. Nat. Immunol. 13, 1118–1128 (2012). NATURE COMMUNICATIONS | (2021) 12:2232 | https://doi.org/10.1038/s41467-021-22536-y | www.nature.com/naturecommunications 11 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-22536-y 32. Gundra, U. M. et al. Alternatively activated macrophages derived from 54. Lauder, C. I., Garcea, G., Strickland, A. & Maddern, G. J. Abdominal adhesion monocytes and tissue macrophages are phenotypically and functionally prevention: still a sticky subject? Dig. Surg. 27, 347–358 (2010). distinct. Blood 123, e110–e122 (2014). 55. Whang, S. H. et al. In search of the best peritoneal adhesion model: 33. Ghosn, E. E. et al. Two physically, functionally, and developmentally distinct comparison of different techniques in a rat model. J. Surg. Res 167, 245–250 peritoneal macrophage subsets. Proc. Natl Acad. Sci. USA 107, 2568–2573 (2011). (2010). 56. Podaru, M. N. et al. Reparative macrophage transplantation for myocardial 34. Wang, J. & Kubes, P. A reservoir of mature cavity macrophages that repair: a refinement of bone marrow mononuclear cell-based therapy. Basic can rapidly invade visceral organs to affect tissue repair. Cell 165, 668–678 Res Cardiol. 114, 34 (2019). (2016). 35. Zhang, N. et al. Expression of factor V by resident macrophages boosts host Acknowledgements defense in the peritoneal cavity. J. Exp. Med 216, 1291–1300 (2019). We thank Professors Sussan Nourshargh and Antal Rot for the critical reading of this 36. Astrup, T. & Stage, A. Isolation of a soluble fibrinolytic activator from animal manuscript, and the members of the Suzuki Laboratory for their valuable discussions. We tissue. Nature 170, 929 (1952). also thank the CRUK Flow Cytometry Core Service at Barts Cancer Institute (Core 37. Hellebrekers, B. W. et al. A role for the fibrinolytic system in postsurgical Award C16420/A18066). This project was funded by the Queen Mary Innovation Proof- adhesion formation. Fertil. Steril. 83, 122–129 (2005). of-Concept fund and British Heart Foundation (Program Grants RG/15/3/31236 and 38. Ryu, J. K. et al. Fibrin-targeting immunotherapy protects against RG/19/7/34577; Project Grant PG/18/77/34100). It was also supported by the UK neuroinflammation and neurodegeneration. Nat. Immunol. 19, 1212–1223 National Institute for Health Research Biomedical Research Center at Barts. (2018). 39. Rosen, H. & Gordon, S. Monoclonal antibody to the murine type 3 complement receptor inhibits adhesion of myelomonocytic cells in vitro Author contributions and inflammatory cell recruitment in vivo. J. Exp. Med 166, 1685–1701 T.I. and K.S. primarily conceived the project and designed the experiments. T.I. per- (1987). formed most experiments with professional supports from Y.S., L.F., M.P., M.S., M.F., 40. Jenkins, S. J. et al. Local macrophage proliferation, rather than recruitment S.K., K.Y., K.K., and F.L. T.I. and K.S. conducted the analysis and interpretation of data. from the blood, is a signature of TH2 inflammation. Science 332, 1284–1288 K.S., T.I., M.P., and F.L. wrote and edited the manuscript with input and approval from (2011). all other authors. 41. Orlofsky, A., Lin, E. Y. & Prystowsky, M. B. Selective induction of the beta chemokine C10 by IL-4 in mouse macrophages. J. Immunol. 152, 5084–5091 (1994). Competing interests The authors declare no competing interests. 42. Nelson, D. S. Reaction to antigens in vivo of the peritoneal macrophages of guinea-pigs with delayed type hypersensitivity. Effects of anticoagulants and other drugs. Lancet 2, 175–176 (1963). Additional information 43. Barth, M. W., Hendrzak, J. A., Melnicoff, M. J. & Morahan, P. S. Review of the Supplementary information The online version contains supplementary material macrophage disappearance reaction. J. Leukoc. Biol. 57, 361–367 (1995). available at https://doi.org/10.1038/s41467-021-22536-y. 44. Gautier, E. L., Ivanov, S., Lesnik, P. & Randolph, G. J. Local apoptosis mediates clearance of macrophages from resolving inflammation in mice. Correspondence and requests for materials should be addressed to T.I. or K.S. Blood 122, 2714–2722 (2013). 45. Daems, W. T. & de Bakker, J. M. Do resident macrophages proliferate? Peer review information Nature Communications thanks the anonymous reviewers for Immunobiology 161, 204–211 (1982). their contribution to the peer review of this work. Peer review reports are available. 46. Bain, C. C. et al. Long-lived self-renewing bone marrow-derived macrophages displace embryo-derived cells to inhabit adult serous cavities. Nat. Commun. Reprints and permission information is available at http://www.nature.com/reprints 7, ncomms11852 (2016). 47. Gundra, U. M. et al. Vitamin A mediates conversion of monocyte-derived Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in macrophages into tissue-resident macrophages during alternative activation. published maps and institutional affiliations. Nat. Immunol. 18, 642–653 (2017). 48. Bain, C. C. et al. Rate of replenishment and microenvironment contribute to the sexually dimorphic phenotype and function of peritoneal macrophages. Open Access This article is licensed under a Creative Commons Sci. Immunol. 5, eabc4466 (2020). Attribution 4.0 International License, which permits use, sharing, 49. Cassado Ados, A., D’Imperio Lima, M. R. & Bortoluci, K. R. Revisiting mouse adaptation, distribution and reproduction in any medium or format, as long as you give peritoneal macrophages: heterogeneity, development, and function. Front appropriate credit to the original author(s) and the source, provide a link to the Creative Immunol. 6, 225 (2015). 50. Takenaka, E., Van Vo, A., Yamashita-Kanemaru, Y., Shibuya, A. & Shibuya, K. Commons license, and indicate if changes were made. The images or other third party Selective DNAM-1 expression on small peritoneal macrophages contributes to material in this article are included in the article’s Creative Commons license, unless CD4(+) T cell costimulation. Sci. Rep. 8, 15180 (2018). indicated otherwise in a credit line to the material. If material is not included in the 51. Gosselin, D. et al. Environment drives selection and function of enhancers article’s Creative Commons license and your intended use is not permitted by statutory controlling tissue-specific macrophage identities. Cell 159, 1327–1340 (2014). regulation or exceeds the permitted use, you will need to obtain permission directly from 52. Nelms, K., Keegan, A. D., Zamorano, J., Ryan, J. J. & Paul, W. E. The IL-4 the copyright holder. To view a copy of this license, visit http://creativecommons.org/ receptor: signaling mechanisms and biologic functions. Annu Rev. Immunol. licenses/by/4.0/. 17, 701–738 (1999). 53. Kamel, R. M. Prevention of postoperative peritoneal adhesions. Eur. J. Obstet. © The Author(s) 2021 Gynecol. Reprod. Biol. 150, 111–118 (2010). 12 NATURE COMMUNICATIONS | (2021) 12:2232 | https://doi.org/10.1038/s41467-021-22536-y | www.nature.com/naturecommunications http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Nature Communications Springer Journals

Loading next page...
 
/lp/springer-journals/cell-barrier-function-of-resident-peritoneal-macrophages-in-post-cEfdXEXY3w

References (68)

Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2021
eISSN
2041-1723
DOI
10.1038/s41467-021-22536-y
Publisher site
See Article on Publisher Site

Abstract

ARTICLE https://doi.org/10.1038/s41467-021-22536-y OPEN Cell barrier function of resident peritoneal macrophages in post-operative adhesions 1 1 1 1 1 1 Tomoya Ito , Yusuke Shintani , Laura Fields , Manabu Shiraishi , Mihai‑Nicolae Podaru , Satoshi Kainuma , 1 1 1 1 1 Kizuku Yamashita , Kazuya Kobayashi , Mauro Perretti , Fiona Lewis-McDougall & Ken Suzuki Post-operative adhesions are a leading cause of abdominal surgery-associated morbidity. Exposed fibrin clots on the damaged peritoneum, in which the mesothelial barrier is dis- rupted, readily adhere to surrounding tissues, resulting in adhesion formation. Here we show High − that resident F4/80 CD206 peritoneal macrophages promptly accumulate on the lesion and form a ‘macrophage barrier’ to shield fibrin clots in place of the lost mesothelium in mice. Depletion of this macrophage subset or blockage of CD11b impairs the macrophage barrier and exacerbates adhesions. The macrophage barrier is usually insufficient to fully preclude the adhesion formation; however, it could be augmented by IL-4-based treatment or adoptive transfer of this macrophage subset, resulting in robust prevention of adhesions. By contrast, monocyte-derived recruited peritoneal macrophages are not involved in the macrophage barrier. These results highlight a previously unidentified cell barrier function of a specific macrophage subset, also proposing an innovative approach to prevent post-operative adhesions. William Harvey Research Institute, Barts and The London School of Medicine and Dentistry, Queen Mary University of London, London, UK. email: t.ito@qmul.ac.uk; ken.suzuki@qmul.ac.uk NATURE COMMUNICATIONS | (2021) 12:2232 | https://doi.org/10.1038/s41467-021-22536-y | www.nature.com/naturecommunications 1 1234567890():,; ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-22536-y 21–25 ost-operative adhesions are pathological connections peritoneal cavity, remains contentious . For instance, clo- binding organs or tissues across a virtual space, e.g., the dronate liposome-induced depletion of peritoneal macrophages Pperitoneal, pericardial or pleural cavity. These are some of attenuates the intra-abdominal adhesion formation in a gauze- the leading causes of post-operative morbidity, resulting in var- induced adhesion model in mice , suggesting a causative role of ious complications . Abdominal adhesions occur in 79–93% of macrophages. On the other hand, adoptive transfer of peritoneal patients who have undergone a major abdominal or pelvic macrophages reduces the adhesion formation in a parietal peri- 2,3 21 procedure . These can cause pain and distress and may result in toneal muscular defect model in rabbits , indicating an anti- more critical complications, i.e., bowel obstruction and female adhesion effect of macrophages. Although the different species infertility . The only effective treatment for post-operative and/or models used in these studies may be a reason for the 5 26 adhesions, once developed, is corrective surgery , whereas this contradicting observations , the heterogeneity and diversity of procedure is invasive and often leads to a recurrence of peritoneal macrophages may also explain this inconsistency. 6 High − adhesions , suggesting the importance of the prevention. The Here we show that resident F4/80 CD206 peritoneal Low + most common preventive approach for post-operative abdominal macrophages, but not recruited F4/80 CD206 macrophages, adhesions is currently implantation or administration of bioma- have an ability to form a “cell barrier” that determines the for- terial barrier products, including artificial films, liquid, or gel, mation of post-operative abdominal adhesions in a mouse model. which keep the injured tissue-surface separated from the neigh- This previously unidentified macrophage barrier is formed via boring tissue/organ . These barrier products are, however, effec- CD11b, but usually insufficient to fully shield the exposed fibrin tive in only half of the patients and have limitations in their clots, thereby allowing the adhesion formation. However, we 8–10 practicability . Therefore, there is a great need for a more describe that IL-4-based treatment augments the macrophage effective and more widely applicable preventative treatment for barrier and results in robust prevention of post-operative adhe- post-operative abdominal adhesions. To this aim, it is essential to sions, proposing an innovative anti-adhesion strategy that is further understand the cellular and molecular mechanism based on a distinctive concept from the conventional treatments. underpinning the adhesion formation post-surgery. The major initiating factor for post-operative adhesion for- Results mation is the production of fibrin clots . Following a surgery- Abdominal adhesion formation within a day of surgery.We associated injury, the coagulation cascade is activated, leading to first investigated the local cellular dynamics related to the the formation of fibrin clots on the surface of damaged tissues . abdominal adhesion formation using a mouse model based on The adhesive polymerization sites on the exposed fibrin clot ischemic button creation on the parietal peritoneum (Supple- 13,14 promote the surrounding tissue to bind with them . The mentary Fig. 1a). The severity of adhesions was evaluated by the endogenous fibrinolysis ability, in which plasmin and plasmino- 4,28 reported adhesion-scoring system . This model demonstrated gen activators play a role, is not so substantial as to attenuate the that the adhesions between the ischemic button and surrounding fibrin clot formation or adhesion formation . Previous studies tissues were formed as early as day 1 post-surgery (Supplemen- have suggested an involvement of various types of cells in this tary Fig. 1b, c). Histological assessments using hematoxylin and 16,17 initial process for post-operative adhesion formation , eosin staining revealed that the mesothelial layer on the ischemic 18–20 including activated mesothelial cells and neutrophils . How- button surface was disrupted by day 1 and that there was an ever, the role of macrophages, which are a major cell type in the accumulation of nucleated cells, which have different Naive (Day 0) Day 1 Day 3 Day 5 (B) (B) (B) (A) (A) (A) (A) Intestine Intestine Intestine Zoom (A) Zoom (A) Zoom (A) Zoom (A) Zoom (B) Zoom (B) Zoom (B) Fig. 1 Post-operative adhesions are formed as early as 1 day after surgery in the mouse ischemic button model. Representative images of hematoxylin- eosin staining of the ischemic button at different time points. Zooms (A) and (B) show the non-adhesion area (A) and adhesion area (B) of the ischemic button, respectively. The inserted images in Zooms (A) present higher magnification views of the black dashed boxes. The naive peritoneum was used as control (Day 0). n = 3 independent experiments. Scale bars represent 2 mm in the lower magnification images and 100 μm in the higher magnification images. 2 NATURE COMMUNICATIONS | (2021) 12:2232 | https://doi.org/10.1038/s41467-021-22536-y | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-22536-y ARTICLE ab Naive (Day 0) Day 1 Day 5 Ischemic button (Day 1) DAPI PDPN F4/80 Peritoneum DAPI PDPN F4/80 Ischemic Peritoneal cavity button Zoom Remote area c d Naive (Day 0) Day 1 Day 3 Day 5 F4/80 cells Peritoneum DAPI PDPN Peritoneal cavity PDPN cells DAPI Fibrin Days after surgery DAPI PDPN F4/80 Fibrin DAPI Button F4/80 DAPI PDPN Intestine F4/80 Fibrin Zoom Zoom Fig. 2 F4/80 macrophages accumulate on fibrin clots in the damaged peritoneum. a Representative tiling image of immunofluorescence staining for the ischemic button on day 1 post-surgery. Cross-sections were stained for PDPN and F4/80. The inserted images present the higher magnification of the surface area of the ischemic button and the remote area from the ischemic button. n = 3 independent ischemic buttons. Scale bar, 1 mm. b Whole-mount immunohistostaining of the ischemic button surface. n = 3 independent ischemic buttons. Scale bars, 100 μm. c, d Representative immunofluorescence images and quantification of ischemic buttons created in the mouse peritoneal membrane. Cross-sections were stained for PDPN, F4/80, and fibrin. The naive peritoneum was examined as “Day 0”. The graph shows the time course of the cover ratio of the ischemic button surface by PDPN mesothelial cells or F4/80 macrophages. n = 6 (Day 0), 5 (Day 1), 7 (Day 3), and 6 (Day 5) mice. Data represent the mean ± SEM. Scale bar, 100 μm. e Representative immunofluorescence images of the adhesion formed between the ischemic button and surrounding tissue (intestine) on day 3 post-surgery. n = 3 independent experiments. Scale bars, 200 μm. morphologies from naive mesothelial cells, on the button surface button was mostly covered by a monolayer of F4/80 macro- (Fig. 1). At the sites of adhesion formation, the mesothelial layer phages instead of PDPN cells (Fig. 2a). We then further pro- was lost by day 1, and connective tissue with cell infiltration was gressed the analysis of every ischemic button by using both formed to bind the ischemic button and the surrounding tissue, whole-mount and cross-section immunohistofluorescent staining. i.e., the intestine. Before the surgical insult, a monolayer of PDPN mesothelial cells with a cobblestone-like morphology completely covered + − the parietal peritoneum, while the majority of these PDPN cells F4/80 CD206 macrophages accumulate on fibrin clots.To disappeared from the button surface by day 1 post-surgery identify the accumulated cells on the ischemic button surface, we (Fig. 2b–d). This disappearance of mesothelial cells was further first performed immunohistofluorescent staining for F4/80 and confirmed by immunostaining for two other mesothelial podoplanin (PDPN) using an ischemic button sample that was markers , cytokeratin 19 and Mesothelin (Supplementary free from adhesion on day 1 after surgery. The result exhibited a Fig. 2a, b). The disrupted mesothelial cell layer was, however, remarkable change in the peritoneal surface of the ischemic reconstructed with PDPN cells by day 5 (Fig. 2b–d). It was button. The parietal peritoneum adjacent and remote to the observed that fibrin clots were formed on the surface of the ischemic button was entirely shielded by the PDPN mesothelial ischemic button with a peak at day 3 (Fig. 2c). Of note, F4/80 cell monolayer, while by contrast the peritoneal surface of the NATURE COMMUNICATIONS | (2021) 12:2232 | https://doi.org/10.1038/s41467-021-22536-y | www.nature.com/naturecommunications 3 Cover ratio of button surface (%) ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-22536-y a b Naive (Day 0) 4 hrs Day 1 Day 3 Day 5 High - 38.6% 3.30% 0.09% 7.99% 22.8% 10 0.6 F4/80 CD206 10 0.4 0.2 0.85% 1.04% 3.04% 8.61% 5.84% Low + F4/80 CD206 01 23 45 3 4 3 4 3 4 3 4 3 4 Days after surgery 010 10 010 10 010 10 010 10 010 10 Anti-CD206-Alexa488 13.0 11.4 c d e ± 0.3 ± 0.2 Day 3 Day 3 Day 5 -6 P = 1.06 x 10 5 *** Ischemic button creation 10 4 High - 10 F4/80 CD206 Low + F4/80 CD206 Liposome DID i.p. injection Low (Day 2 after surgery) F4/80 CD206 + 5 F4/80 CD206 - - F4/80 CD206 Analysis for peritoneal cavity cells 3 4 3 4 5 3 4 5 010 10 010 10 10 010 10 10 (Day 3 or 5 after surgery) Anti-CD206-Alexa488 Liposome DID Fig. 3 Two subsets of peritoneal macrophages exhibit different cellular properties and distinct dynamics post-surgery. a, b Representative flow High − Low + cytometry plots (a) and cell counts (b) of F4/80 CD206 macrophages (red) and F4/80 CD206 (gray) macrophages in the peritoneal fluid after ischemic button creation in mice. n = 7 (Naive; Day 0), 6 (4 h), 8 (Day 1), 8 (Day 3) and 9 (Day 5) mice. See Supplementary Fig. 5 for the gating strategy. Data are shown as the mean ± SEM. c Schematic of the phagocytosis uptake assay. d Representative flow cytometry histogram showing the phagocytosis High − Low + Low − activity of peritoneal F4/80 CD206 macrophages (red) and F4/80 CD206 (gray) macrophages in comparison with F4/80 CD206 (orange), − + − − F4/80 CD206 (green) and F4/80 CD206 (blue) cells in the peritoneal cavity. n = 3 independent experiments. e Diameter of sorted peritoneal F4/ High − Low + 80 CD206 and F4/80 CD206 macrophages. Data are presented as box-and-whisker plots (interquartile ranges (IQRs) as boxes, with the median High − as a black line and the whiskers extending up to the most extreme points within 1.5-fold IQR). n = 109 (F4/80 CD206 ) cells and 154 (F4/ Low + 80 CD206 ) cells. ***P < 0.001, two-tailed Student’s t-test. High − macrophages accumulated on these lesions and appeared to activity (Fig. 3c, d). Of note, while F4/80 CD206 macro- substitute for the disrupted mesothelial barrier albeit in part phages were dominant (~40% of the total peritoneal fluid cells) in (50–65% of the surface area covered; Fig. 2d). These macrophages the peritoneal fluid in the normal state, they were almost cleared were localized particularly on the surface of exposed fibrin clots, to occupy only 1% of peritoneal fluid cells by 4 h post-surgery consequently shielding ~60% of the surface area of exposed fibrin (Fig. 3a, b). This clearance was more evident at 24 h post-surgery, clots on day 3 (Supplementary Fig. 2c) and showed a phagocy- at which these cells were hardly detectable. We speculated that High − tosis ability (Supplementary Fig. 3a, b). This macrophage barrier this loss of F4/80 CD206 macrophages from the fluid was was temporal and cleared by day 5 when the mesothelial barrier due to the firm attachment of these cells to fibrin clots on the was reconstituted. ischemic button as described in the histological investigation + High − We found that F4/80 macrophages accumulating on the (Fig. 2a–d). The F4/80 CD206 macrophage subset was button surface were negative for CD206 (Supplementary Fig. 4). mostly reconstituted in the peritoneal fluid by day 5. On the other + Low + By contrast, CD206 cells, which were weakly positive for F4/80, hand, F4/80 CD206 macrophages were rarely found in the were observed within the ischemic button tissue after day 3, naive peritoneal fluid, but became detectable, occupying 3% of the indicating a distinction between these two macrophage subsets. peritoneal fluid cells, on day 1 with a peak at day 3. Throughout Low + Furthermore, at the lesion where an abdominal adhesion had the time course studied, the number of F4/80 CD206 mac- High − been formed between the ischemic button and surrounding tissue, rophages was similar to or smaller than that of F4/80 CD206 reduced accumulation of F4/80 macrophages on the fibrin clots macrophages (Fig. 3b). Both of these peritoneal macrophage was noted (Fig. 2e). These results collectively suggest that the cell subsets exhibited high expressions of tissue repair-related genes + − barrier composed of F4/80 CD206 macrophages on the and low expressions of pro-inflammatory or angiogenesis-related exposed fibrin clots may have the potential to contribute to genes (Supplementary Fig. 6a, b). As compared to the other High − attenuation of post-operative adhesion formation. subset, F4/80 CD206 macrophages were larger in size and exhibited higher mRNA or protein expression of Gata6, Tgfb2, Peritoneal macrophage subsets have distinct dynamics.We Ucp1, and ICAM2 (Fig. 3e and Supplementary Fig. 7a, b), which 30–32 then characterized the surface phenotype of peritoneal macro- are reported markers for resident peritoneal macrophages . Low + phages during the course of post-operative abdominal adhesion By contrast, the smaller F4/80 CD206 macrophages exhibited formation by using flow cytometry. Consistent with the above higher CCR2 and MHCII expression (Supplementary Fig. 7b), histological findings, two major macrophage subpopulations, F4/ suggesting their identity as monocyte-derived, recruited High − Low + 32,33 80 CD206 and F4/80 CD206 subsets, were identified macrophages . (Fig. 3a, b, and Supplementary Fig. 5). Both of these macrophage To examine the possibility that the other recruited monocytes/ Low − − + subsets, but not F4/80 CD206 , F4/80 CD206 or F4/80 macrophage subsets or granulocytes have the ability to cover the − − fibrin clot, we have further characterized the peritoneal cavity CD206 peritoneal cells, showed an evident phagocytosis 4 NATURE COMMUNICATIONS | (2021) 12:2232 | https://doi.org/10.1038/s41467-021-22536-y | www.nature.com/naturecommunications High - F4/80 CD206 Low + F4/80 CD206 Anti-F4/80-PE Anti-F4/80-PE Cell number (x10 ) Cell diameter (µm) NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-22536-y ARTICLE cells after ischemic button creation. The results showed that F4/ macrophages. This suggests that the fibrin clot itself has the Low − High 80 CD206 cells included Ly6C classical monocytes/ ability to attract resident peritoneal macrophages. Low macrophages, Ly6C non-classical monocytes/macrophages, We hypothesized that CD11b (integrin alpha M), which is + + 38 and Siglec-F eosinophils, while Ly6G neutrophils were found known to regulate the binding of microglia and fibrin , might − − in F4/80 CD206 population (Supplementary Fig. 8a, b). play a role in this interaction between resident peritoneal However, these cell types did not accumulate on the button macrophages and fibrin clots. A high-level CD11b expression of High − surface (Supplementary Fig. 9a). Furthermore, accumulated resident F4/80 CD206 peritoneal macrophages was con- F4/80 macrophages on fibrin clots were positive for a resident firmed (Supplementary Fig. 12b). Furthermore, proximity ligation macrophage marker, ICAM2 (Supplementary Fig. 9b, c). None- assay (PLA) detected clear PLA signals between fibrin and CD11b theless, there remained a concern that recruited monocytes/ on F4/80 macrophages (Fig. 5c), but not between fibrin and F4/ macrophages or granulocytes would not cover the fibrin clot 80 (Supplementary Fig. 12c). Furthermore, intraperitoneal because of their too-small occurrence at the time of initial injection of a blocking antibody to CD11b (clone; 5C6) adhesion formation (<24 h post-surgery). Therefore, we further intensified the abdominal adhesion, which corresponded to the assessed the ability of recruited monocytes/macrophages to abolished accumulation of F4/80 macrophages on the fibrin clot participate in the cell barrier formation in the “repeated ischemic (Fig. 5d–g). In addition, the above-mentioned transient disap- High − button creation model” (Supplementary Fig. 10a, b). In this pearance of resident F4/80 CD206 peritoneal macrophages model, we created an additional ischemic button on the right-side from the peritoneal fluid 1 day after surgery (Fig. 3a, b) was peritoneal wall on day 3 after the first ischemic button creation on inhibited by blocking CD11b, while the occurrence of recruited Low + the left-side peritoneal wall. This model enabled us to investigate F4/80 CD206 peritoneal macrophages in the peritoneal fluid the capability of recruited monocytes/macrophages more pre- was unaffected (Supplementary Fig. 12d, e). Thus, we conclude High − cisely because the occurrence of these recruited cells in the that accumulation of resident F4/80 CD206 peritoneal peritoneal cavity is much more abundant at day 3 compared to macrophages on the fibrin clot exposed on the damaged the time period for the ordinary cell barrier formation. The peritoneum was achieved directly via CD11b. obtained result demonstrated that, even with such an increased occurrence, recruited monocytes/macrophages or granulocytes did not contribute to the cell barrier formation. These results IL-4 reduces adhesions by enhancing macrophage barriers. together confirm that only resident macrophages have the ability Having identified the cell barrier function of resident F4/ High − to cover the fibrin clots. 80 CD206 peritoneal macrophages, this naturally-occurring process was not sufficient in preventing abdominal adhesions (Fig. 1 and Supplementary Fig. 1b, c). We thus explored if forced Depletion of resident macrophages exacerbates adhesions.We High − augmentation of resident F4/80 CD206 peritoneal macro- next investigated the functional role of resident peritoneal mac- phages would result in robust prevention of post-operative rophages in the post-operative adhesion formation by using a adhesion formation. We confirmed that intraperitoneal admin- 34,35 depletion strategy. Based on the previous reports , we adjusted istration of long-acting IL-4 complex (IL-4c) increased the the protocol of pre-injection of clodronate liposomes so that High − number of resident F4/80 CD206 peritoneal macrophages High − resident F4/80 CD206 peritoneal macrophages were depleted approximately three times (Fig. 6a, b). This IL-4c treatment at the time of surgery (Fig. 4a). This protocol did not reduce the indeed resulted in a markedly attenuated formation of post- Low + increase of recruited F4/80 CD206 peritoneal macrophages operative adhesions with increased coverage of the exposed fibrin post-surgery (Supplementary Fig. 11a, b). As a result of this clots by F4/80 macrophages (Fig. 6c–f). Histological assessments clodronate liposome treatment, we found markedly exacerbated demonstrated that mesothelial cells on the peritoneal membrane abdominal adhesions after ischemic button creation with a or fibrin deposition on the ischemic button lesion were unaffected reduced cover rate of fibrin clots by F4/80 macrophages by IL-4c injection (Fig. 6g–i). In addition, IL-4c administration (Fig. 4b–f). Of importance, the degree of the mesothelial dis- had no effect on other types of immune cells in the peritoneal ruption or the amount of fibrin deposition was unaffected cavity or within the ischemic button tissue (Supplementary (Fig. 4g–i and Supplementary Fig. 11c), suggesting that the Fig. 13a–c). Supporting this observation, IL-4c administration did High − depletion of resident F4/80 CD206 peritoneal macrophages not affect the Th1/Th2-related cytokine production levels in the did not affect fibrin formation or degradation. Consistent with peritoneal fluid 3 days after surgery, while a macrophage-derived this, the main pro- or anti-fibrinolytic factor, tissue plasminogen chemokine, CCL6/C10 , was up-regulated (Supplementary 36 37 activator (tPA) or plasminogen activator inhibitor-1 (PAI-1) , Fig. 14a, b). in the peritoneal fluid was not affected by the clodronate lipo- Consistent with the previous report that IL-4 promotes not some treatment (Fig. 4j, k). These data indicate that resident F4/ only resident macrophage proliferation but also recruited High − 80 CD206 peritoneal macrophages contribute to the macrophage polarization , IL-4c administration also increased attenuation of post-operative abdominal adhesion formation Low + the number of recruited F4/80 CD206 macrophages in the through reducing the exposure of fibrin clots as a cell barrier, but peritoneal fluid (Fig. 6a, b), while this cell number was less than not through affecting fibrin production or fibrinolysis. High − one-third of the F4/80 CD206 macrophage number. To Low + exclude the possibility that recruited F4/80 CD206 macro- Resident macrophages bind fibrin clots via CD11b. To under- phages contributed to the attenuation of post-operative adhe- stand the mechanism by which the fibrin clot attracted resident sions, we isolated each macrophage subsets from IL-4c injected peritoneal macrophages, we implanted an ex vivo produced fibrin mice and performed adoptive transfer into the peritoneal cavity of clot using a TISEEL fibrin glue kit (Baxter) into the mouse a syngeneic recipient mouse at the time of the ischemic button peritoneal cavity (Supplementary Fig. 12a). As a result, it was creation. As we expected, adoptive transfer of resident F4/ High − observed that a large number of resident peritoneal macrophages 80 CD206 peritoneal macrophages, but not recruited F4/ Low + accumulated onto the surface of the implanted exogenous fibrin 80 CD206 peritoneal macrophages, significantly reduced clot, similar to the finding of the fibrin clot on the ischemic post-operative adhesions (Fig. 6j, k). Accumulation of a sizeable High − button (Fig. 5a, b). A more than half of the exogenous fibrin clot number of transferred resident F4/80 CD206 peritoneal surface was covered by the accumulated resident peritoneal macrophages on the surface of the ischemic button was NATURE COMMUNICATIONS | (2021) 12:2232 | https://doi.org/10.1038/s41467-021-22536-y | www.nature.com/naturecommunications 5 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-22536-y b d PBS-Lipo Clod-Lipo PBS-Lipo Clod-Lipo -5 P = 5.23 x 10 *** DAPI 1.0 F4/80 0.8 Fibrin 0.6 -3 P = 5.05 x 10 0.4 ** 0.2 3.0 2.0 1.0 Zoom n.s. ef h j -7 n.s. P = 1.04 x 10 2.0 *** 100 1.5 DAPI PDPN F4/80 Fibrin 75 n.s. 1.0 50 0.5 Button Peritoneal cavity Day 0 Day 1 gi k n.s. n.s. 6.0 10.0 8.0 n.s. 4.0 6.0 50 n.s. 4.0 2.0 2.0 0 0 Day 0 Day 1 High − Fig. 4 Depletion of resident F4/80 CD206 peritoneal macrophages exacerbates post-operative abdominal adhesion formation. a The cell number High − of resident F4/80 CD206 peritoneal macrophages in the peritoneal cavity on day 2 after intraperitoneal injection of clodronate liposomes compared to the control PBS-liposome injection. n = 6 mice in each group. b, c The visual appearance of the abdominal adhesion formation (b) and the adhesion score (c) on day 1 post-ischemic button creation with an intraperitoneal injection of clodronate liposomes or PBS-liposomes two days before surgery. n = 15 mice in each group. d Representative images of the whole-mount immunohistostaining of the ischemic button on day 1 after surgery with pre-injection of clodronate liposomes or PBS-liposomes. n = 3 mice in each group. Scale bars, 100 μm. e–i Representative cross-section immunofluorescent images and quantification of the ischemic button on day 1 after surgery with the injection of clodronate liposomes or PBS-liposomes. The bar graphs show the coverage + + of exposed fibrin clots by F4/80 cells (f) or PDPN cells (g). The fibrin area (h) or thickness (i) on the ischemic button surface were also histologically quantified. n = 6 (PBS-Lipo group) and 7 (Clod-Lipo group) mice. Scale bars, 100 μm. j, k ELISA measurements for active tPA protein (j) or active PAI-1 protein (k) in the peritoneal fluid after surgery with pre-injection of clodronate liposomes or PBS-liposomes. n = 4 (Day 0) and 5 (Day 1) mice in each group. Data represent the mean ± SEM. **P < 0.01, ***P < 0.001, ns not significant, two-tailed student’s t-test. confirmed (Fig. 6l). These data proposed that the IL-4-based macrophages on their surface through CD11b-mediated binding. treatment would achieve substantial prevention of post-operative This macrophage accumulation results in the formation of a cell adhesion formation through augmentation of the resident barrier that acts to shield the adhesive fibrin clot from the sur- peritoneal macrophage barrier that shields exposed fibrin clots. rounding tissue in place of the lost mesothelial barrier, potentially attenuating the adhesion formation. The macrophage barrier is formed promptly by day 1 post-surgery and is temporal being replaced by the reconstructed mesothelial cell barrier by day 5. Discussion Compared to the complete mesothelial barrier, the macrophage Despite its enormous clinical importance, cellular and molecular barrier is insufficient; it covers only ~60% of the exposed fibrin processes underlying the formation and/or prevention of post- clot area. As a result, post-operative adhesions are developed at operative adhesions remains poorly understood . Here, in a the lesions that have an inadequate macrophage barrier. Of mouse model, we reveal a previously unidentified role of resident clinical impact, the resident macrophage barrier can be aug- peritoneal macrophages to attenuate post-operative adhesion mented by IL-4-based treatment or adoptive cell transfer, formation through the generation of a cell barrier (Fig. 7). Shortly which leads to robust prevention of the adhesion formation, after surgical insult, sticky fibrin clots are formed on the damaged proposing an innovative anti-adhesion strategy. In addition, peritoneal membrane, where the mesothelial barrier is lost. High − recruiting peritoneal macrophages are not involved in the These fibrin clots capture resident F4/80 CD206 peritoneal 6 NATURE COMMUNICATIONS | (2021) 12:2232 | https://doi.org/10.1038/s41467-021-22536-y | www.nature.com/naturecommunications PBS-Lipo Clod-Lipo PBS-Lipo Clod-Lipo PBS-Lipo Clod-Lipo PBS-Lipo PBS-Lipo Clod-Lipo Clod-Lipo PBS-Lipo Clod-Lipo Clod-Lipo PBS-Lipo High - F4/80 CD206 cell number (x10 ) Adhesion score Fibrin cover by Fibrin cover by PDPN cells (%) + F4/80 cells (%) Fibrin thickness Fibrin area on (µm) button surface (%) Active PAI-1 (ng/ml) Active tPA (ng/ml) PBS-Lipo PBS-Lipo Clod-Lipo Clod-Lipo PBS-Lipo PBS-Lipo Clod-Lipo Clod-Lipo NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-22536-y ARTICLE a Surface of b Cross-section of exogenous fibrin clot Zoom exogenous fibrin clot Exogenous fibrin clot DAPI DAPI F4/80 F4/80 Fibrin Fibrin Peritoneal cavity -2 P = 3.14 x 10 cd Anti-CD11b e IgG2b PLA 3.0 Fibrin CD11b (Fibrin / CD11b) 2.0 Button 1.0 Peritoneal cavity DAPI F4/80 Fibrin g PLA Fibrin -2 F4/80 Zoom P = 2.47 x 10 Button CD11b F4/80 IgG2b Peritoneal cavity Anti- CD11b High − Fig. 5 Resident F4/80 CD206 peritoneal macrophages bind fibrin clots via CD11b. a, b Representative confocal image of the whole-mount (a) and cross-section (b) immunofluorescent staining of the implanted exogenous fibrin clot on day 1 after implantation. n = 3 mice. Scale bars, 100 μm. (c) Representative confocal images of PLA between CD11b of F4/80 macrophages and fibrin on the ischemic button surface. PLA signals are shown in red. n = 3 independent experiments. Scale bars, 5 μm. d, e Representative images of abdominal adhesions (d) and adhesion scores (e) on day 1 post-ischemic button creation with an intraperitoneal injection of control IgG2b antibody (IgG2b group, n = 6 mice) or anti-CD11b blocking antibody (Anti-CD11b group, n = 7 mice). f, g Representative cross-section immunohistostaining images of the ischemic button (f) and quantification of the coverage of exposed fibrin clots by F4/80 cells (g) on day 1 after surgery with an intraperitoneal injection of control IgG2b antibody or anti-CD11b blocking antibody. n = 5 mice in each group. Scale bar, 100 μm. Data represent the mean ± SEM. *P < 0.05, two-tailed Student’s t-test. protective cell barrier mechanism, shedding light on a previously disappearance reaction (MDR) in the setting of intra-abdominal 42,43 unknown biological insight regarding the complex diversity of infection . It is also known that intraperitoneal injection of 30 35 44 macrophages. lipopolysaccharide , zymosan , or thioglycolate induces the We here demonstrate that the accumulation of resident peri- MDR and that the MDR is suppressed by pretreatment with 35,42 toneal macrophages on the injured peritoneum is mediated anticoagulants, including warfarin and heparin . The under- through binding between CD11b and fibrin. There is a report pinning mechanism of the MDR is thought to be macrophage describing that peritoneal macrophages are attracted to the accumulation to a foreign body or migration into the lymphatic 35,43 damaged tissue through sensing extracellular ATP and contribute vessels, rather than the death of macrophages . Our study to the removal of dead cells in a heat-induced liver injury model reports that the MDR occurs during post-operative adhesion 34 High − in mice . However, we observe that implantation of an formation as well. Resident F4/80 CD206 peritoneal mac- externally-produced fibrin clot, which is unlikely to release a rophages that occupy around 40% of the total peritoneal fluid significant level of ATP, results in macrophage accumulation on cells in the normal status are almost cleared as early as day 1 after its surface, similar to that observed in the ischemic button model. surgery. The series of histological examinations in this study Moreover, in addition to the positive PLA signals between CD11b suggest that this clearance from the peritoneal fluid is due to and fibrin, the inhibition of CD11b-fibrin interaction using a the attachment of these cells to the fibrin clots on the ischemic blocking antibody abolishes the accumulation of resident mac- button. By day 5 after surgery, the population of resident High − rophages onto the damaged peritoneum of the ischemic button. F4/80 CD206 peritoneal macrophages is reconstituted in the Taken together, although a possible role of extracellular ATP and peritoneal fluid. Although the source of these reconstituting other factors cannot be fully excluded, we conclude that the macrophages remains uncertain, it could be their detachment CD11b-fibrin binding is a major mechanism by which resident from the fibrin clots, migration from the peritoneal organs , peritoneal macrophages accumulated onto the injured perito- differentiation from recruited monocytes, and/or phenotype Low + neum in our model of post-operative adhesion. change from recruited F4/80 CD206 peritoneal macrophages 46–48 One of the interesting findings in this study is that resident or other cells types . peritoneal macrophages drastically but transiently disappeared Peritoneal macrophages are classified into two major distinct from the peritoneal fluid after adhesion-inducing surgery. A subsets, the large peritoneal macrophage (LPM) and small peri- similar phenomenon has been reported as the macrophage toneal macrophage (SPM), basically according to their size . NATURE COMMUNICATIONS | (2021) 12:2232 | https://doi.org/10.1038/s41467-021-22536-y | www.nature.com/naturecommunications 7 IgG2b Anti-CD11b IgG2b Anti-CD11b Fibrin cover by Adhesion score F4/80 cells (%) ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-22536-y a b PBS IL-4c -2 P = 1.91 x 10 10 2.0 2.0 -3 P = 2.18 x 10 1.5 3 1.5 ** 1.0 1.0 0.5 0.5 3 4 3 4 010 10 010 10 0 PBS IL-4c PBS IL-4c Anti-CD206-Alexa488 c ef h -2 P = 2.43 x 10 PBS IL-4c PBS IL-4c n.s. Button 100 DAPI 75 F4/80 50 50 25 25 Peritoneal cavity 0 0 DAPI PBS IL-4c -2 PBS IL-4c P = 2.34 x 10 PDPN g i n.s. F4/80 3.0 Fibrin 100 15 2.0 n.s. 1.0 50 Zoom 25 0 0 PBS IL-4c PBS IL-4c PBS IL-4c j l High - Low + PBS F4/80 CD206 F4/80 CD206 Day 1 Day 3 Day 5 Button DAPI F4/80 CM-DiI Fibrin Peritoneal cavity n.s. -2 P = 3.32 x 10 3.0 Zoom 2.0 1.0 Fig. 6 IL-4c administration augments the macrophage barrier and attenuates post-operative adhesion formation. a, b Representative flow cytometry plots (a) and cell numbers (b) of peritoneal macrophage subsets day 2 after injection of IL-4c or PBS only. n = 5 mice in each group. c, d The representative appearance of post-operative adhesions (c) and the adhesion score (d) on day 3 post-surgery. IL-4c or PBS was administered into the peritoneal cavity immediately after ischemic button creation surgery. n = 9 mice per group. e–i Representative immunofluorescence images and quantification of cross- sections of the ischemic button on day 3 after surgery with IL-4c or PBS administration. The bar graphs show the coverage of exposed fibrin clots by F4/80 + + cells (f) or PDPN cells (g). The fibrin area (h) or thickness (i) on the ischemic button surface were also histologically measured. n = 5 (PBS) and 7 (IL- 4c) mice. Scale bars, 100 μm. j, k Representative images of post-operative abdominal adhesions (j) and calculated adhesion scores (k) on day 5 after High − Low + ischemic button creation with intraperitoneal adoptive transfer of syngeneic mouse F4/80 CD206 or F4/80 CD206 peritoneal macrophages, or High − Low + PBS administration. n = 5 (PBS), 4 (F4/80 CD206 ) and 5 (F4/80 CD206 ) mice. l Representative immunofluorescence images of the ischemic High − button with adoptive transfer of resident F4/80 CD206 peritoneal macrophages (pre-labeled with CM-DiI; red). n = 2 independent experiments. Scale bar, 100 μm. Data represent the mean ± SEM. *P < 0.05, **P < 0.01, ns not significant, one-way ANOVA and Tukey’s post hoc test in (k) or two-tailed student’s t-test in others. Low Low High Low LPMs are the resident macrophages, which are dominant at the exhibit an F4/80 , CD11b , MHCII , GATA6 , and steady-state and maintained in the peritoneal cavity through self- CD206 phenotype. Sexually dimorphic replenishment of LPM High High Low renewal, and have an F4/80 , CD11b , MHCII , by bone marrow-derived cells, which is high in males and very High − 46,49 48 GATA6 , and CD206 phenotype . By contrast, SPMs are low in females, has also been reported . In addition, there is the recruited macrophages, which are derived from circulating information available regarding the functional diversity of these monocytes and increase in number in response to an inflam- subsets. For instance, LPMs regulate the peritoneal B-1 cell- matory stimulus in a CCR2 dependent manner . These cells mediated gut IgA production through TGFβ2 under the normal 8 NATURE COMMUNICATIONS | (2021) 12:2232 | https://doi.org/10.1038/s41467-021-22536-y | www.nature.com/naturecommunications PBS High - F4/80 CD206 Low + F4/80 CD206 Anti-F4/80-PE Adhesion score Adhesion score High - F4/80 CD206 cell number (x10 ) Low + F4/80 CD206 Fibrin cover by Fibrin cover by Cell number (x10 ) + + PDPN cells (%) F4/80 cells (%) Fibrin area on Fibrin thickness the button surface (%) (µm) NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-22536-y ARTICLE Surgical trauma Insufficient macrophage barrier Mesothelial loss & Fibrin clot still exposed Adhesion fibrin clot formation formation Sufficient macrophage barrier Mesothelial cells High - F4/80 CD206 macrophages + IL-4c Fibrin clot Fibrin clot covered by macrophages Re-mesothelialization Fig. 7 The function of macrophage barriers during adhesion formation. The schematic illustrates the role of resident macrophage barriers during post- High − operative abdominal adhesion formation. Upon injury of the peritoneal membrane, resident F4/80 CD206 peritoneal macrophages promptly accumulate onto the lesion where fibrin clots are exposed. These accumulated macrophages form an anti-adhesion cell barrier to shield the sticky fibrin clots from neighboring tissues. This macrophage barrier is, however, usually insufficient, allowing the formation of adhesions (Upper panels). However, when this cell barrier is sufficiently strengthened i.e., using IL-4c treatment, post-operative adhesions are robustly prevented (Lower panels). condition . SPMs have an ability to present antigens to naive responses to IL-4 and the resulting effects on adhesion formation + 50 CD4 T cells , suggesting that SPMs induce local T-cell priming could be elucidated by using cell-type-specific IL-4 receptor during peritoneal infection or inflammation. In our study, knockout mice. Of note, a previous study has reported that IL-4 High − Low + 24 F4/80 CD206 and F4/80 CD206 macrophage subsets knockout did not affect post-operative adhesions in mice , exhibited comparable characters to LPMs and SPMs, respectively, indicating that endogenous IL-4 is unlikely to play a critical role 30,33,51 in terms of morphology and gene/protein expression . in the adhesion formation. However, further investigation is needed to define these macro- The findings obtained in this study have significant clinical phages as LPMs and SPMs. implications in developing an innovative anti-adhesion strategy. + + We observed that sizeable numbers of CD11b CCR2 To prevent post-operative adhesion formation, a range of phar- High + + Ly6C classical monocytes/macrophages and CD11b CCR2 macological agents, including fibrinolytic agents, non-steroidal Low Ly6C non-classical monocytes/macrophages, which were F4/ anti-inflammatory drugs, steroids, antioxidants, histamine Low − 80 CD206 , appeared in the peritoneal cavity at day 3 post- antagonists, and statins, were tested; however, none of them has High − Low + 53 surgery. Unlike F4/80 CD206 and F4/80 CD206 mac- been clinically successful to date . Instead, implantation/ Low − rophage subsets, these F4/80 CD206 cell populations did not administration of a biomaterial barrier product is currently present a phagocytic ability, suggesting that they were unlikely to conducted for the prevention of adhesions . Nonetheless, this be mature macrophages. Furthermore, our results showed that method has obvious limitations in its efficacy and + 8–10 CCR2 recruited monocytes/macrophages did not have the practicability . Macrophage barriers are likely to have multiple ability to form an anti-adhesion cell barrier. On the other hand, important advantages over the biomaterial-based barrier. The there is a report showing that pre-treatment of rabbits with formation of a macrophage barrier relies on a natural, physiolo- proteose peptone or transplantation of proteose peptone-elicited gical healing process, while an artificial biomaterial barrier is peritoneal macrophages protect against adhesion formation, associated with possible foreign body reactions . The macro- suggesting that monocyte-derived inflammatory macrophages phage barrier has no risk of physical displacement, unlike a 21 9,10 might offer an anti-adhesion effect in a certain condition .Itis biomaterial barrier . In addition, the mobility and flexibility of speculated that this effect of monocyte-derived macrophages may macrophages enable the formation of a cell barrier at any location be attained through a different mechanism(s) from cell barrier throughout the peritoneal cavity, including difficult-to-access formation, including modulation of inflammation, fibrin forma- localizations, i.e., deep inside of the pelvic cavity or between tion, fibrinolysis and/or vacuolization, and fibrosis of adhesion intestines. Furthermore, sufficient macrophage barriers can be 21–25 tissues . achieved simply by injection of a pharmacological reagent, i.e., IL- IL-4 is known to play pleiotropic roles in the immune system . 4c, and is thereby accessible in any type of surgical procedures, We found that intraperitoneal IL-4c administration attenuated including open, laparoscopic, or robotic surgeries. Further post-operative adhesion formation by increasing the number of development of this innovative treatment to prevent post- High − F4/80 CD206 resident peritoneal macrophages. Adoptive operative adhesions, which is based on the previously unknown High − transfer of IL-4c-treated F4/80 CD206 macrophages resulted concept of the macrophage barrier, is warranted. To this end, it is in comparable attenuation of adhesions. On the other hand, IL-4c essential to confirm the universality of the results obtained in this administration did not affect the number of non-macrophage cell study. We here used the mouse ischemic button model because it types including lymphocytes or the major Th1/Th2-related has an important advantage in inducing reproducible intra- cytokine levels in the peritoneal fluid. Therefore, we consider that abdominal adhesions compared to other models as previously the resident peritoneal macrophage barrier is the central demonstrated . Indeed, this model offered reliable and adequate mechanism for the anti-adhesion effect of IL-4c administration. adhesion formation to our study. As the next step toward clinical However, we cannot exclude a possibility of functional alterations application, investigations using a different model, particularly a of these cells in response to IL-4, which would modulate a process large animal model, and also research with human cells will be of adhesion formation. Further detailed cell-type dependent required. NATURE COMMUNICATIONS | (2021) 12:2232 | https://doi.org/10.1038/s41467-021-22536-y | www.nature.com/naturecommunications 9 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-22536-y Methods Peritoneal cell isolation and flow cytometry analysis. After mice were sacrificed Study approval. All investigations using living animals conformed to the Princi- using CO inhalation, peritoneal cells were obtained by washing the peritoneal cavity with ice-cold PBS four times. Erythrocytes were removed by incubating with ples of Laboratory Animal Care formulated by the National Society for Medical Research and to NIH guidelines (Guide for the Care and Use of Laboratory Ani- red blood cell lysis buffer (Biolegend, 420301) at 5 min on ice. Total live cell number was counted using a Countess II Automated Cell Counter (Invitrogen, mals. National Academies Press 1996) and performed with the approval of the ethics committee of the Queen Mary University of London and the UK Home AMQAX1000) with Trypan Blue staining. After centrifugation, peritoneal cells were Office (Project License PPL70/8503). All in vivo procedures were performed by UK resuspended in FACS buffer (HBSS plus 2 mM EDTA and 0.5% BSA) and pre- Home Office Personal License holders. In vivo and in vitro procedures and incubated with an anti-mouse CD16/CD32 antibody (clone: 93) (1:100 dilution, assessments were blinded wherever possible. Invitrogen, 14–0161–85) to block the Fc receptor. Dead cells and debris were excluded by forwarding scatter/side scatter (FSC/SSC) and DAPI staining (1:500 dilution). Phenotype analysis was performed by staining peritoneal cells with the fluorophore-conjugated antibodies for 3 h at 4 °C. Information about primary Post-operative abdominal adhesion model. Ten- to twelve-week-old male antibodies and dilutions used in the assay can be found in “Supplementary Table 1”. C57BL/6 mice were purchased from Charles River Laboratories UK and used in the In order to accurately identify the positive signal, appropriate isotype antibodies experiments. These mice were maintained in a specific pathogen-free room under a were used as a negative control. Flow cytometric analyses were performed using a 12-h light/12-dark cycle, 20–24 °C temperature, and 45–55% humidity with free BD LSRFortessa cell analyzer (BD Biosciences) with FlowJo software version 10 access to food and water in our animal facility. Surgery was performed under (Tree Star). Exemplifying the gating strategy is provided in the Supplementary aseptic conditions, and all efforts were made to minimize animal suffering. Information (Supplementary Fig. 5). Cell sorting was performed using a FACS Aria Anesthesia was induced by 2.0% isoflurane inhalation. Abdominal adhesion for- High − II (BD Biosciences). To measure the cell size of individual F4/80 CD206 and mation was induced by the construction of two ischemic buttons on the peritoneal Low + F4/80 CD206 macrophages, the digital images of sorted cells were acquired with wall, as previously described . After a small median laparotomy, two ischemic an All-in-One microscope (BZ-8000; KEYENCE) using a 20× objective lens (1360 × buttons, ~5 mm in diameter each and spaced ~10 mm apart, were created on the 1024 pixels). Cell diameter was determined using ImageJ software. right-side peritoneal wall by clamping and ligating with a 5–0 Mersilk suture (W2500T, Ethicon). The abdomen was closed with a double-layered suture of the peritoneum and skin with a 5–0 vicryl absorbable suture (W9915, Ethicon). They Whole-mount staining. Tissues or exogenous fibrin clots were fixed in 4% PFA were allowed to recover on a heating pad and monitored daily until euthanasia. At overnight at 4 °C and permeabilized and blocked in PBS containing 0.5% Triton X- the chosen time point, the mice were sacrificed with carbon dioxide (CO ) inha- 100 and 5% BSA for 4 h at 4 °C. Subsequently, samples were labeled with primary lation to evaluate abdominal adhesion formation, and peritoneal fluid/tissue antibodies overnight at 4 °C. After rinsing in PBS three times for 5 min, the sections 4,28 samples collected. According to the adhesion scoring system , the adhesion were next incubated with DAPI and the appropriate fluorophore-conjugated sec- scores for each ischemic button (Lower; tail side, Upper; head side) were deter- ondary antibodies overnight at 4 °C. Confocal images were captured in z-stacks of mined as follows: 0, no adhesion; 1, filmy adhesion (separated easily by blunt 9–13 planes and reconstructed with the maximum intensity projection using ZEN dissection); 2, firm adhesion (separated by aggressive blunt dissection); 3, dense black 3.0 lite software (Carl Zeiss). All antibodies using whole-mount staining are adhesion (separable only by sharp dissection). Adhesion scores were validated in a described in Supplementary Table 1. blinded manner, and the average of lower and upper ischemic button scores was calculated (Supplementary Fig. 1a). For the repeated ischemic button creation Administration of liposomes, blocking antibodies and IL-4c. For macrophage model (Supplementary Fig. 10a), two ischemic buttons were first created on the depletion, 50 μg of clodronate liposomes (Clod-Lipo) or control liposomes (PBS- left-side peritoneal wall of a mouse as described above. Three days after this first Lipo) (Encapsula NanoSciences, CLD-8901) were injected intraperitoneally 2 days surgery, the abdomen was re-opened, and an additional ischemic button was before surgery. In order to determine the phagocytosis ability of peritoneal cavity created on the right-side peritoneal wall. cells, 50 μg of Fluoroliposome-DiD (Encapsula Nano Sciences, CLD-8913) were injected intraperitoneally 2 days after surgery. For blocking macrophage-fibrin interaction, 50 μg anti-CD11b neutralizing antibody (clone: 5C6) (Invitrogen, MA5- Hematoxylin and eosin staining and immunohistolabeling. After harvesting, 16528) or rat IgG2b control antibody (clone: eB149/10H5; eBioscience,16–4031–85) tissues were fixed in 4% paraformaldehyde (PFA) overnight at 4 °C and were was injected intraperitoneally 3 h before and immediately after surgery. For embedded and frozen in O.C.T. compound (VWR International). increasing macrophages, IL-4c (5 μg recombinant IL-4 (PeproTech, 214–14) and 25 For H&E staining of ischemic buttons and adhesion tissues, 6 µm thick frozen μg stabilizing monoclonal anti-IL-4 antibody (clone: BVD4-1D11; BD Biosciences, sections were cut and incubated in Mayer’s hematoxylin solution (Sigma-Aldrich, 554387)) dissolved in 100 μl PBS were intraperitoneally injected to normal mice or MHS16) for 15 min and Eosin Y solution (Sigma-Aldrich, HT110116) for 1 min. administrated into the peritoneal cavity immediately after the ischemic button After dehydration through increasing concentrations of ethanol to xylene, the creation. An equivalent volume of PBS was injected as a control. sections were mounted using the DPX mounting medium (VWR International, 13512). The digital images were acquired with an All-in-One microscope (BZ-8000; KEYENCE) using a 2× or 20× objective lens (1360 × 1024 pixels). Active tPA and PAI-1 measurements. Peritoneal fluid from the adhesion model For immunohistolabeling, frozen sections were cut at 6 µm thick and pre- mice that were pre-injected with clodronate liposomes or control liposomes was blocked with the blocking buffer (PBS plus 5% BSA and 0.1% Tween 20) for 30 collected through washing the peritoneal cavity with 500 μl ice-cold PBS. Protein min at room temperature. To block endogenous biotin, if required, samples were concentrations of the peritoneal fluid were determined using a DC Protein Assay pre-blocked with streptavidin buffer (PBS plus 0.1 mg/ml Streptavidin and 0.1% (Bio-Rad, 5000112). ELISA was performed according to the manufacturer’s Tween 20) for 30 min and biotin buffer (PBS plus 0.5 mg/ml Biotin and 0.1% instructions of mouse active tPA functional assay ELISA kit (Innovative Research, Tween-20) for 30 min at room temperature. Samples were labeled with the IMSTPAKTA) and active mouse PAI-1 functional assay ELISA kit (Innovative primary antibodies overnight at 4 °C. Information about primary antibodies and Research, IMSPAI1KTA). dilutions used in the assay can be found in “Supplementary Table 1”.The anti- laminin antibody (1:1000 dilution, Sigma, L9393) was conjugated with Alexa Exogenous fibrin clot implantation into the mouse abdomen.Exogenous fibrin Fluor 488 fluorophores using labeling kits (Life Technologies, A20181) clots for implantation were produced using TISSEEL (Baxter, 1502243). Briefly, 10 μl according to the manufacturer’s instruction. After rinsing in PBS three times for of component 1 (Human fibrinogen, 72–110 mg/ml; Aprotinin (synthetic), 3000 5 min, the sections were next incubated with 4′,6-diamidino-2-phenylindole KIU/ml) and 10 μl of component 2 (Human thrombin, 500 IU/ml; Calcium chloride (DAPI) (Sigma-Aldrich, D9542) and the appropriate fluorophore-conjugated dihydrate, 40 µmol/ml) were mixed on a sliding glass using a pipette tip. For easy secondary antibodies or AlexaFluor 546-conjugated Streptavidin (1:300 dilution, identification of exogenous fibrin clots in histological analyses, 10 mg/ml Alexa Fluor Invitrogen, S11225) for 1 h at the room temperature. Stained sections were 647-conjugated human fibrinogen (Invitrogen, F35200) was mixed in component 1 mounted with DAKO Fluorescence Mounting Medium (Agilent, S302380-2), (1:500 dilution). After fibrin clot formation, the exogenous clots (~5 mm in dia- and the digital images were acquired, and pseudo-colored with an All-in-One meter) were washed five times with PBS to remove excess salt and implanted into the microscope (BZ-8000; KEYENCE) using a 10× or 20× objective lens (1360 × peritoneal cavity of a mouse through a laparotomy (Supplementary Fig. 12a). 1024 pixels). All image analysis was performed by importing images as TIFF files into ImageJ version 1.50i software (NIH). To visualize a whole image of the ischemic button Proximity ligation assay. Interaction between CD11b and fibrin was examined (Fig. 2a), sixteen images were reconstructed by using the MosaicJ plugin. For the using reagents from the Duolink proximity ligation assay (Sigma-Aldrich, + + quantification of the cover ratio by F4/80 macrophages or PDPN mesothelial DUO92106) following the manufacturer’s instructions. Briefly, 4% PFA-fixed cells on the ischemic button and fibrin clots, F4/80 or PDPN positive area was frozen sections were cut at 6 µm thick and pre-blocked with the blocking buffer quantified as the percentage of the surface of peritoneal area or fibrin positive area. (PBS plus 5% BSA and 0.1% Tween 20) for 30 min at room temperature. Samples For quantification of fibrin formation, the fibrin area and thickness were quantified were labeled with anti-CD11b antibody (clone: M1/70; 1:200 dilution, eBioscience, as the percentage or average thickness of the fibrin positive area on the ischemic 14–0112–85) and anti-fibrin antibody (1:500 dilution, Abcam, ab34269) overnight button (Supplementary Fig. 11c). Images containing the surface area of the at 4 °C. After rinsing in PBS three times for 5 min, the sections were incubated with ischemic button, excluded adhesion area, were acquired at three independent Goat AlexaFluor 488-conjugated anti-rat IgG antibody (1:300 dilution, Invitrogen, regions and analyzed. A color threshold function of ImageJ was applied to measure A11006) for 1 h at room temperature. Slides were washed and incubated with a macrophages, mesothelial cells, and fibrin-clot area. PLA probe antibody, which is conjugated with PLUS or MINUS short nucleotide 10 NATURE COMMUNICATIONS | (2021) 12:2232 | https://doi.org/10.1038/s41467-021-22536-y | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-22536-y ARTICLE sequences, for 1 h at 37 °C. PLUS and MINUS oligonucleotides were ligated and 2. Menzies, D. & Ellis, H. Intestinal obstruction from adhesions-how big is the amplified by rolling circle PCR with fluorescent nucleotides for 90 min at 37 °C. problem? Ann. R. Coll. Surg. Engl. 72,60–63 (1990). After rolling circle amplification, slides were washed and incubated with APC-anti- 3. Stommel, M. W. J. et al. Multicenter observational study of adhesion F4/80 antibody (clone: BM8; 1:200 dilution, Biolegend, 123116) and Goat Alexa- formation after open-and laparoscopic surgery for colorectal cancer. Ann. Fluor 405-conjugated anti-rabbit IgG antibody (1:300 dilution, Invitrogen, A31556) Surg. 267, 743–748 (2018). for 1 h at room temperature. F4/80 and fibrin interaction was examined as a 4. Ten Broek, R. P. G. et al. Bologna guidelines for diagnosis and management of negative control, in the same manner, using anti-F4/80 antibody (clone: BM8; adhesive small bowel obstruction (ASBO): 2017 update of the evidence-based 1:200 dilution, eBioscience, 14–4801–81) and APC-anti-CD11b antibody (clone: guidelines from the world society of emergency surgery ASBO working group. M1/70; 1:200 dilution, eBioscience, 17–0112–82). PLA was visualized using an World J. Emerg. Surg. 13, 24 (2018). inverted Zeiss 800 (Carl Zeiss) confocal laser scanning microscope equipped with 5. Sutton, C. Adhesions and their prevention. Obstet. Gynaecol. 7, 168–176 63× (1.4 NA) objective and solid-state laser diodes (405, 488, 561, and 640 nm (2005). excitation wavelengths). 6. Wiseman, D. M., Trout, J. R. & Diamond, M. P. The rates of adhesion development and the effects of crystalloid solutions on adhesion development Adopted macrophage transfer. Peritoneal cavity cells were collected from four IL- in pelvic surgery. Fertil. Steril. 70, 702–711 (1998). 7. Stapleton, L. M. et al. Use of a supramolecular polymeric hydrogel as an 4c administrated, syngeneic donor mice on day 5 post-ischemic button creation. High − Low + F4/80 CD206 and F4/80 CD206 macrophages were isolated using a FACS effective post-operative pericardial adhesion barrier. Nat. Biomed. Eng. 3, Aria II (BD Biosciences), as described above. We transferred 400,000 cells in 100 μl 611–620 (2019). PBS per recipient mouse directly into the peritoneal cavity immediately after sur- 8. Ten Broek, R. P. G. et al. Benefits and harms of adhesion barriers for gery. After 5 days, the mice were sacrificed and adhesion scores evaluated. For abdominal surgery: a systematic review and meta-analysis. Lancet 383,48–59 High − enabling histological tracing of transferred macrophages, isolated F4/80 CD206 (2014). peritoneal macrophages were labeled with CM-DiI before transplantation (Invi- 9. Brochhausen, C. et al. Current strategies and future perspectives for trogen, C7000) according to the manufacturer’s recommendations. intraperitoneal adhesion prevention. J. Gastrointest. Surg. 16, 1256–1274 (2012). 10. Trochsler, M. & Maddern, G. J. Adhesion barriers for abdominal surgery: a sticky problem. Lancet 383,8–10 (2014). RNA extraction and real-time PCR. Total RNA was extracted using the PicoPure™ High − 11. Koninckx, P. R., Gomel, V., Ussia, A. & Adamyan, L. Role of the peritoneal RNA Isolation Kit (Applied Biosystems, KIT0204) from F4/80 CD206 and Low + cavity in the prevention of postoperative adhesions, pain, and fatigue. Fertil. F4/80 CD206 macrophages, which were isolated from the peritoneal cavity Steril. 106, 998–1010 (2016). 5 days post-surgery using a cell sorter as described above. The RNA concentration 12. Mutsaers, S. E., Prele, C. M., Pengelly, S. & Herrick, S. E. Mesothelial cells and and quality were measured using the DS-11 Spectrophotometer (DeNovix). cDNA peritoneal homeostasis. Fertil. Steril. 106, 1018–1024 (2016). was prepared using the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, 4368814) according to the manufacturer’s instruction. Real-time PCR 13. diZerega, G. S. & Campeau, J. D. Peritoneal repair and post-surgical adhesion was performed by Quant studio 7 qPCR machine (Applied Biosystems) with a formation. Hum. Reprod. Update 7, 547–555 (2001). PowerUP SYBR Green Master Mix (Applied Biosystems, A25742). Bone marrow 14. Maciver, A. H., McCall, M. & James Shapiro, A. M. Intra-abdominal mononuclear cells (BMMNCs) were collected from the femurs and tibiae of normal adhesions: cellular mechanisms and strategies for prevention. Int J. Surg. 9, mice as previously described and were used as a control. Fold changes were 589–594 (2011). calculated by relative quantification based on the second derivative method. Gene 15. Hellebrekers, B. W. & Kooistra, T. Pathogenesis of postoperative adhesion expression levels were normalized by the Gapdh gene. All primers are described in formation. Br. J. Surg. 98, 1503–1516 (2011). “Supplementary Table 2”. 16. Kosaka, H., Yoshimoto, T., Yoshimoto, T., Fujimoto, J. & Nakanishi, K. Interferon-gamma is a therapeutic target molecule for prevention of postoperative adhesion formation. Nat. Med 14, 437–441 (2008). Cytokine profiling. The peritoneal fluid was collected through washing the peri- 17. Braun, K. M. & Diamond, M. P. The biology of adhesion formation in the toneal cavity with 500 μl ice-cold PBS of the mice on day 3 post-ischemic button peritoneal cavity. Semin Pediatr. Surg. 23, 336–343 (2014). creation surgery with IL-4c or PBS administration. After determining the protein 18. Tsai, J. M. et al. Surgical adhesions in mice are derived from mesothelial cells concentration using a DC Protein Assay (Bio-Rad, 5000112), the peritoneal fluid and can be targeted by antibodies against mesothelial markers. Sci. Transl. containing 20 μg total protein was pooled from 6 mice (total 120 μg protein for Med 10, eaan6735 (2018). each group). Cytokine profiling was performed according to the manufacturer’s instructions of Mouse XL Cytokine Array Kit (R&D Systems, ARY028). Briefly, 19. Tsai, J. M. et al. Neutrophil and monocyte kinetics play critical roles in mouse peritoneal adhesion formation. Blood Adv. 3, 2713–2721 (2019). following the block with blocking buffer at room temperature for 1 h, the mem- branes were incubated with pooled peritoneal fluids overnight at 4 °C. After rinsing 20. Fischer, A. et al. Post-surgical adhesions are triggered by calcium-dependent with washing buffer three times for 10 min, membranes were incubated with a membrane bridges between mesothelial surfaces. Nat. Commun. 11, 3068 biotinylated detection antibody cocktail at room temperature for 1 h and then (2020). incubated with streptavidin–horseradish peroxidase for 30 min. Membranes were 21. Ar’Rajab, A. et al. Enhancement of peritoneal macrophages reduces exposed to the peroxidase substrate and measured with the Azure c600 Imager postoperative peritoneal adhesion formation. J. Surg. Res 58, 307–312 (1995). (Azure Biosystems). The images were analyzed using ImageJ software. 22. Burnett, S. H. et al. Development of peritoneal adhesions in macrophage depleted mice. J. Surg. Res 131, 296–301 (2006). 23. Hoshino, A. et al. Inhibition of CCL1-CCR8 interaction prevents aggregation Statistics. All statistical tests were performed using RStudio version 1.0.153 soft- of macrophages and development of peritoneal adhesions. J. Immunol. 178, ware and GraphPad Prism 8 software (GraphPad Software). Data represent the 5296–5304 (2007). mean ± SEM. For comparisons between multiple groups, one-way ANOVA was 24. Hong, G. S. et al. Effects of macrophage-dependent peroxisome proliferator- performed, followed by Tukey’s post hoc test. Two groups were compared using activated receptor gamma signalling on adhesion formation after abdominal the two-tailed, unpaired Student’s t-test. surgery in an experimental model. Br. J. Surg. 102, 1506–1516 (2015). 25. Honjo, K. et al. Plasminogen activator inhibitor-1 regulates macrophage- Reporting summary. Further information on research design is available in the Nature dependent postoperative adhesion by enhancing EGF-HER1 signaling in mice. Research Reporting Summary linked to this article. FASEB J. 31, 2625–2637 (2017). 26. Beyene, R. T., Kavalukas, S. L. & Barbul, A. Intra-abdominal adhesions: Anatomy, physiology, pathophysiology, and treatment. Curr. Probl. Surg. 52, Data availability 271–319 (2015). Data that support the findings of this study are available within the article and Supplementary Information or from the corresponding author upon reasonable request. 27. Buckenmaier, C. C. 3rd, Pusateri, A. E., Harris, R. A. & Hetz, S. P. Comparison Raw source data are present in the Source Data file Source data are provided with of antiadhesive treatments using an objective rat model. Am. Surg. 65, 274–282 (1999). this paper. 28. Coccolini, F. et al. Peritoneal adhesion index (PAI): proposal of a score for the “ignored iceberg” of medicine and surgery. World J. Emerg. Surg. 8, 6 (2013). Received: 5 August 2020; Accepted: 15 March 2021; 29. Lua, I., Li, Y., Pappoe, L. S. & Asahina, K. Myofibroblastic conversion and regeneration of mesothelial cells in peritoneal and liver fibrosis. Am. J. Pathol. 185, 3258–3273 (2015). 30. Okabe, Y. & Medzhitov, R. Tissue-specific signals control reversible program of localization and functional polarization of macrophages. Cell 157, 832–844 References (2014). 1. Krielen, P. et al. Adhesion-related readmissions after open and laparoscopic 31. Gautier, E. L. et al. Gene-expression profiles and transcriptional regulatory surgery: a retrospective cohort study (SCAR update). Lancet 395,33–41 pathways that underlie the identity and diversity of mouse tissue (2020). macrophages. Nat. Immunol. 13, 1118–1128 (2012). NATURE COMMUNICATIONS | (2021) 12:2232 | https://doi.org/10.1038/s41467-021-22536-y | www.nature.com/naturecommunications 11 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-22536-y 32. Gundra, U. M. et al. Alternatively activated macrophages derived from 54. Lauder, C. I., Garcea, G., Strickland, A. & Maddern, G. J. Abdominal adhesion monocytes and tissue macrophages are phenotypically and functionally prevention: still a sticky subject? Dig. Surg. 27, 347–358 (2010). distinct. Blood 123, e110–e122 (2014). 55. Whang, S. H. et al. In search of the best peritoneal adhesion model: 33. Ghosn, E. E. et al. Two physically, functionally, and developmentally distinct comparison of different techniques in a rat model. J. Surg. Res 167, 245–250 peritoneal macrophage subsets. Proc. Natl Acad. Sci. USA 107, 2568–2573 (2011). (2010). 56. Podaru, M. N. et al. Reparative macrophage transplantation for myocardial 34. Wang, J. & Kubes, P. A reservoir of mature cavity macrophages that repair: a refinement of bone marrow mononuclear cell-based therapy. Basic can rapidly invade visceral organs to affect tissue repair. Cell 165, 668–678 Res Cardiol. 114, 34 (2019). (2016). 35. Zhang, N. et al. Expression of factor V by resident macrophages boosts host Acknowledgements defense in the peritoneal cavity. J. Exp. Med 216, 1291–1300 (2019). We thank Professors Sussan Nourshargh and Antal Rot for the critical reading of this 36. Astrup, T. & Stage, A. Isolation of a soluble fibrinolytic activator from animal manuscript, and the members of the Suzuki Laboratory for their valuable discussions. We tissue. Nature 170, 929 (1952). also thank the CRUK Flow Cytometry Core Service at Barts Cancer Institute (Core 37. Hellebrekers, B. W. et al. A role for the fibrinolytic system in postsurgical Award C16420/A18066). This project was funded by the Queen Mary Innovation Proof- adhesion formation. Fertil. Steril. 83, 122–129 (2005). of-Concept fund and British Heart Foundation (Program Grants RG/15/3/31236 and 38. Ryu, J. K. et al. Fibrin-targeting immunotherapy protects against RG/19/7/34577; Project Grant PG/18/77/34100). It was also supported by the UK neuroinflammation and neurodegeneration. Nat. Immunol. 19, 1212–1223 National Institute for Health Research Biomedical Research Center at Barts. (2018). 39. Rosen, H. & Gordon, S. Monoclonal antibody to the murine type 3 complement receptor inhibits adhesion of myelomonocytic cells in vitro Author contributions and inflammatory cell recruitment in vivo. J. Exp. Med 166, 1685–1701 T.I. and K.S. primarily conceived the project and designed the experiments. T.I. per- (1987). formed most experiments with professional supports from Y.S., L.F., M.P., M.S., M.F., 40. Jenkins, S. J. et al. Local macrophage proliferation, rather than recruitment S.K., K.Y., K.K., and F.L. T.I. and K.S. conducted the analysis and interpretation of data. from the blood, is a signature of TH2 inflammation. Science 332, 1284–1288 K.S., T.I., M.P., and F.L. wrote and edited the manuscript with input and approval from (2011). all other authors. 41. Orlofsky, A., Lin, E. Y. & Prystowsky, M. B. Selective induction of the beta chemokine C10 by IL-4 in mouse macrophages. J. Immunol. 152, 5084–5091 (1994). Competing interests The authors declare no competing interests. 42. Nelson, D. S. Reaction to antigens in vivo of the peritoneal macrophages of guinea-pigs with delayed type hypersensitivity. Effects of anticoagulants and other drugs. Lancet 2, 175–176 (1963). Additional information 43. Barth, M. W., Hendrzak, J. A., Melnicoff, M. J. & Morahan, P. S. Review of the Supplementary information The online version contains supplementary material macrophage disappearance reaction. J. Leukoc. Biol. 57, 361–367 (1995). available at https://doi.org/10.1038/s41467-021-22536-y. 44. Gautier, E. L., Ivanov, S., Lesnik, P. & Randolph, G. J. Local apoptosis mediates clearance of macrophages from resolving inflammation in mice. Correspondence and requests for materials should be addressed to T.I. or K.S. Blood 122, 2714–2722 (2013). 45. Daems, W. T. & de Bakker, J. M. Do resident macrophages proliferate? Peer review information Nature Communications thanks the anonymous reviewers for Immunobiology 161, 204–211 (1982). their contribution to the peer review of this work. Peer review reports are available. 46. Bain, C. C. et al. Long-lived self-renewing bone marrow-derived macrophages displace embryo-derived cells to inhabit adult serous cavities. Nat. Commun. Reprints and permission information is available at http://www.nature.com/reprints 7, ncomms11852 (2016). 47. Gundra, U. M. et al. Vitamin A mediates conversion of monocyte-derived Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in macrophages into tissue-resident macrophages during alternative activation. published maps and institutional affiliations. Nat. Immunol. 18, 642–653 (2017). 48. Bain, C. C. et al. Rate of replenishment and microenvironment contribute to the sexually dimorphic phenotype and function of peritoneal macrophages. Open Access This article is licensed under a Creative Commons Sci. Immunol. 5, eabc4466 (2020). Attribution 4.0 International License, which permits use, sharing, 49. Cassado Ados, A., D’Imperio Lima, M. R. & Bortoluci, K. R. Revisiting mouse adaptation, distribution and reproduction in any medium or format, as long as you give peritoneal macrophages: heterogeneity, development, and function. Front appropriate credit to the original author(s) and the source, provide a link to the Creative Immunol. 6, 225 (2015). 50. Takenaka, E., Van Vo, A., Yamashita-Kanemaru, Y., Shibuya, A. & Shibuya, K. Commons license, and indicate if changes were made. The images or other third party Selective DNAM-1 expression on small peritoneal macrophages contributes to material in this article are included in the article’s Creative Commons license, unless CD4(+) T cell costimulation. Sci. Rep. 8, 15180 (2018). indicated otherwise in a credit line to the material. If material is not included in the 51. Gosselin, D. et al. Environment drives selection and function of enhancers article’s Creative Commons license and your intended use is not permitted by statutory controlling tissue-specific macrophage identities. Cell 159, 1327–1340 (2014). regulation or exceeds the permitted use, you will need to obtain permission directly from 52. Nelms, K., Keegan, A. D., Zamorano, J., Ryan, J. J. & Paul, W. E. The IL-4 the copyright holder. To view a copy of this license, visit http://creativecommons.org/ receptor: signaling mechanisms and biologic functions. Annu Rev. Immunol. licenses/by/4.0/. 17, 701–738 (1999). 53. Kamel, R. M. Prevention of postoperative peritoneal adhesions. Eur. J. Obstet. © The Author(s) 2021 Gynecol. Reprod. Biol. 150, 111–118 (2010). 12 NATURE COMMUNICATIONS | (2021) 12:2232 | https://doi.org/10.1038/s41467-021-22536-y | www.nature.com/naturecommunications

Journal

Nature CommunicationsSpringer Journals

Published: Apr 14, 2021

There are no references for this article.