Leukocyte-Associated Immunoglobulin-like Receptor-1 is regulated in human myocardial infarction but its absence does not affect infarct size in mice

Heart failure after myocardial infarction (MI) depends on infarct size and adverse left ventricular (LV) remodelling, both influenced by the inflammatory response. Leukocyte-associated immunoglobulin-like receptor 1 (LAIR-1) is an inhibitory receptor of ITAM-dependent cell activation, present on almost all immune cells. We investigated regulation of LAIR-1 leukocyte expression after MI in patients and hypothesized that its absence in a mouse model of MI would increase infarct size and adverse remodelling. In patients, LAIR-1 expression was increased 3 days compared to 6 weeks after MI on circulating monocytes (24.8 ± 5.3 vs. 21.2 ± 5.1 MFI, p = 0.008) and neutrophils (12.9 ± 4.7 vs. 10.6 ± 3.1 MFI, p = 0.046). In WT and LAIR-1−/− mice, infarct size after ischemia-reperfusion injury was comparable (37.0 ± 14.5 in WT vs. 39.4 ± 12.2% of the area at risk in LAIR-1−/−, p = 0.63). Remodelling after permanent left coronary artery ligation did not differ between WT and LAIR-1−/− mice (end-diastolic volume 133.3 ± 19.3 vs. 132.1 ± 27.9 μL, p = 0.91 and end-systolic volume 112.1 ± 22.2 vs. 106.9 ± 33.5 μL, p = 0.68). Similarly, no differences were observed in inflammatory cell influx or fibrosis. In conclusion, LAIR-1 expression on monocytes and neutrophils is increased in the acute phase after MI in patients, but the absence of LAIR-1 in mice does not influence infarct size, inflammation, fibrosis or adverse cardiac remodelling.


Patient data collection.
In patients, at the moment of inclusion (3 days after MI) and 6 weeks thereafter, venous blood was drawn and collected by the Laboratory of Clinical Chemistry and Haematology of the UMC Utrecht (Fig. 1a). Whole blood was directly subjected to flow cytometry and EDTA plasma was stored at −80° Celsius in the UMC Utrecht Biobank and used for ELISA (see below). Similarly, blood was drawn from healthy controls and subjected to flow cytometry.
Animals. Healthy male C57Bl/6 LAIR-1 −/−20 and C57Bl/6 WT littermates (age 10-12 weeks, weight 25-30 g) were housed at room temperature under 12 hour light/dark cycle in groups of maximum 5 animals (in type III cages with filtertop) under strict DM1 regulations and received standard chow and water ad libitum. Healthy controls and patients with a first-time MI were included in the current study. Venous blood collection in healthy controls and at 3 days and 6 weeks after MI was used for flow cytometric analyses and to collect plasma (a). Mice were either subjected to ischemia-reperfusion injury or to permanent left coronary artery ligation. In the mice subjected to 30 minutes of ischemia and 24 hours of subsequent reperfusion, IS/AAR staining was performed (b). In the mice subjected to permanent left coronary artery ligation and sacrificed after 3 days, we performed flow cytometric analyses on various tissues and blood (c). In the other mice subjected to left coronary artery ligation, we performed echocardiography at baseline, 7 days and 28 days after MI and performed similar flow cytometric analyses (d -lower panel). MI: myocardial infarction; IS/AAR: infarct size/area at risk.
SCIENTIfIC REPORTS | (2017) 7:18039 | DOI: 10.1038/s41598-017-13678-5 All animals were genotyped prior to the experimental procedure and animal welfare was monitored daily. A blinded researcher performed surgery on randomly assigned animals (random number generation in excel to animal number, which resulted in alternating fashion of operation of WT and LAIR-1 −/− mice). Blinded technicians and observers performed the respective operations, data acquisition and analyses. Sample size calculation for myocardial ischemia reperfusion model was based on infarct size as the primary endpoint. With a power of 90%, alpha of 0.05, estimated effect size of 13% difference in infarct size, standard deviation of 10.4% (based on historical data) and estimated peri-operative mortality of 5% this resulted in a group size of 15 animals per group. For the permanent ligation model sample size calculation was based on end systolic volume as the primary endpoint. With a power of 90%, alpha of 0.05, estimated effect size of 20 μl difference in volume, standard deviation of 15 μl (based on historical data) and estimated peri-operative mortality of 25% this resulted in a group size of 20 animals per group. Cellular influx and collagen content in the ischemic area were defined as secondary outcomes. All animal experiments were approved by the Ethical Committee on Animal Experimentation of the University Medical Center Utrecht (Utrecht, the Netherlands) and conform to the 'Guide for the care and use of laboratory animals' .
Induction of myocardial ischemia-reperfusion injury. The experimental set-up and timeline of myocardial ischemia-reperfusion injury is displayed in Fig. 1b. All operations were performed in the morning before noon. In a dedicated mouse operation room, anaesthesia was induced by intraperitoneal (i.p.) injection of medetomidinehydrochloride (1.0 g/kg body weight), midazolam (10.0 mg/kg) and fentanyl (0.1 mg/kg). These anaesthetics were preferred over cardioprotective propofol or volatile anaesthetics (e.g. isoflurane) 21 . Mice were intubated and connected to a respirator with a 1:1 oxygen-air ratio (times/minute). A core body temperature of 37 °C was maintained during surgery by continuous rectal temperature monitoring and an automatic heating blanket. The heart was accessed through a left lateral thoracotomy with incision of the pericardium. The left coronary artery was ligated for 30 minutes with an 8-0 Ethilon suture (Ethicon) with a section of polyethylene-10 tubing placed over the left coronary artery (LCA). Ischemia was confirmed by bleaching of myocardium and tachycardia. After 30 minutes of ischemia, reperfusion was initiated by releasing the ligation, resulting in tissue colour recurrence. A piece of the suture was left in place to allow for accurate ligature positioning and determination of the ischemic area and the area at risk at termination. The surgical wounds were closed and subcutaneous atipamezole hydrochloride (3.3 mg/kg), flumazenil (0.5 mg/kg) and buprenorphin (0.15 mg/kg) were used as an antagonist. The evening of the day of operation and every 12 hours thereafter, subcutaneous injection of buprenorphin (0.15 mg/kg) was administered as analgesia.
Infarct size and area at risk quantification after ischemia-reperfusion injury. In total 30 animals (15 WT and 15 LAIR-1 −/− ) were subjected to ischemia-reperfusion injury. Twenty-four hours after ischemia-reperfusion injury, mice were euthanized using sodium pentobarbital (60.0 mg/kg) and a left re-thoracotomy was performed. The LCA was ligated at the same location as it was ligated during index ischemia. The thoracic aorta was cannulated and 2% Evans blue was injected upstream in the aorta to perfuse the coronaries, allowing for staining of the remote but not the area at risk (AAR). The heart was then explanted and rinsed with 0.9% saline to remove superfluous dye. The left ventricle (LV) was dissected and a small piece of gauze was inserted in the left ventricular cavity. After one hour at −20 °C, the LV was cut into 4 equally sized sections. Sections were placed in 1% 2,3,5-triphenyltetrazolium chloride (TTC) in saline and incubated at 37 °C for 20 minutes. After 10 minutes, sections were turned to allow for adequate reagent contact. Then, sections were placed in formalin and photographs of both sides of each tissue section were captured using a SZH10 Olympus Zoom Stereo Microscope and IC Capture software, version 2.4. The infarct (white), border zone (red) and remote area (blue) were quantified using ImageJ (version 1.48 v). Infarct size (IS) was expressed as a percentage of the AAR and as a percentage of the LV. Induction of myocardial infarction by permanent ligation. Permanent coronary artery ligation was performed as described above for ischemia-reperfusion injury, but leaving the ligature in place, resulting in a permanent occlusion of the left coronary artery. The experimental set-up including the timeline of the mice sacrificed after either 3 days or 28 days is shown in Fig. 1c,d. Surgery was performed on 23 animals (10 WT and 13 LAIR-1 −/− ) for 3 day survival and 38 animals (18 WT and 20 LAIR-1 −/− ) were included for long term survival (28 days).

Survival.
Mice that died after MI were thoroughly inspected for the cause of death. Deaths within 48 hours after MI were considered due to perioperative complications or direct complication of MI. Cardiac rupture was confirmed by massive intrathoracic haemorrhage >48 hours after operation and ventricular leakage of the myocardium upon perfusion of the heart with 0.9% saline. In the ischemia reperfusion model 2 LAIR-1 −/− died due to perioperative conditions. Of the animals exposed to 3 days permanent ligation 1 WT animal died during the surgical procedure, of animals exposed to 28 days permanent ligation 6 WT and 9 LAIT-1 −/− mice died. Of the 6 WT, 5 died due to cardiac rupture and 1 to unkown causes. Of the LAIR-1 −/− 8 died due to cardiac ruputre and 1 to unknown causes.
Echocardiography. At baseline, 7 and 28 days after permanent ligation, anaesthesia was induced by inhalation of 2.0% isoflurane in a mixture of oxygen/air (1:1). Echocardiography was used to assess cardiac geometry and function. Heart rate, respiration and rectal temperature were constantly monitored and body temperature was kept between 36.0 and 38.0 °C using heat lamps. Respiration gating, a 3-dimensional motor and trigger points were used to obtain 300 transversal images of the heart during the expiratory phase, either at the end of systole or the end of diastole. These images were then used for complete 3D reconstruction of the heart. Image acquisition and analyses were performed using the dedicated Vevo ® 2100 System and Software (Fujifilm VisualSonics Inc., Toronto, Canada). Tissue processing and histological analyses. At the end of the follow-up period, mice were euthanized using sodium pentobarbital (60.0 g/kg). Blood was collected through orbital puncture in EDTA tubes. The inferior caval vein was incised and the vascular system was flushed with 5 mL phosphate-buffered saline (PBS) through right ventricular puncture.
Neutrophils, T-cells, macrophages and collagens were semi-automatically quantified using digital histology. Collagen content was quantified in tissue sections stained for picrosirius red and photographed under polarized light, converted to gray scale images and expressed as a percentage of the region of interest (i.e. infarct, remote). Images of tissue sections were captured and analysed using CellSens (Olympus Corporation, Tokyo, Japan). Of the 9 WT and 10 LAIR-1 −/− ) mice at 3 days follow-up, 1 WT and 1 LAIR −/− mice were excluded for histological analysis due to the absence of a clearly identifiable infarction. For neutrophil analysis an 2 WT and 2 LAIR-1 −/− ) animals were excluded due to technical errors. Of the 11 WT and 10 LAIR1 −/− mice at 28 days follow-up 1 WT and 1 LAIR-1 −/− ) mice were excluded for histological analysis due to the absence of a clearly identifiable infarction. For macrophage analysis an 3 LAIR-1 −/− ) animals were excluded due to technical errors. 2. Technical errors may occur due to poor quality upon sectioning or inferior quality of immunohistochemistry stainings, which makes reliable analysis impossible.
Flow cytometric assays. Fresh human EDTA blood (50 μL) was added to an antibody mixture containing different cell surface markers to identify neutrophils and monocytes (see Supplementary Table S1). Cells were incubated for 30 minutes in the dark at room temperature (RT). Before measurement, cells were washed and erythrocytes were lysed using Optilyse C.
To harvest single cells from heart tissue, enzymatic degradation was performed (N = 6 WT and N = 6 LAIR-1 −/− ). Infarct and remote tissue were collected 3 days after MI and cut into small pieces of around 1 mm 2 . Dissociation solution (10 × 10 2 U/ml DNase I (Roche 04536282001), 10 mM HEPES (Life Technology 15630-080) and 2.6 U/ml Liberase TL (Roche 05401020001)) was added to the tissue and incubated at 37° Celsius for 20 minutes. Single cells of the dissociated myocardial tissue, lymph nodes and spleen were obtained through gentle filtering over a 40 µm cell strainer and subsequently incubated with an antibody mixture containing different cell surface markers to identify neutrophils, monocytes, and T-and B-lymphocytes (see Supplementary Table S2) for 30 minutes in the dark at RT. After washing, residual red blood cells were lysed with erythrocyte-lysis buffer. All samples were measured on a Gallios flow cytometer (10 colour configuration, Beckman Coulter, Marseille, France). Kaluza Analysis Software 1.3 was used for data analysis. The gating strategy is shown in Supplementary Fig. S1.
ELISA. Plasma levels of soluble LAIR-1 (sLAIR-1) and LAIR-2 were measured in duplo using a respective sLAIR-1 and LAIR-2 sandwich ELISA according to manufacturer's instructions (LifeSpan BioSciences, Seatle, WA, USA). Colorimetric analyses were performed using a spectrophotometer (450 nm). Plasma levels were calculated based on standards.
Statistical analyses. Data distribution was evaluated for normality using the d' Agostino & Pearson normality test. Data are expressed as mean ± standard deviation (SD). Skewed ELISA and immunoassay data were ln-transformed and presented as median with interquartile range (IQR). Normally distributed data were compared using a two-tailed paired (serial measurements) or unpaired t-test (separate groups). Non-normally distributed data were compared using a Wilcoxon (serial measurements) or Mann-Whitney test (separate groups). A log-rank (Mantel-Cox) test was used for survival analysis. A level of p < 0.05 was considered statistically significant. Statistical analyses were performed using SPSS software, version 21 and GraphPad Prism, version 6.
Plasma levels of sLAIR-1 were slightly higher 3 days after MI compared to 6 weeks, though this was not significant (2. Fig. 2h).
Survival, cardiac geometry, and cardiac function are comparable between wild-type and LAIR-1 −/− mice after permanent coronary artery ligation. Within

Wild-type and LAIR −/− mice show no differences in inflammatory responses following myocardial infarction.
To confirm LAIR-1 expression on circulating leukocytes, we performed flow cytometry on baseline blood. In WT mice, LAIR-1 was expressed on CD4 + T-cells and CD8 + T-cells, but most prominent  LAIR-1 expression on leukocyte subsets and sLAIR-1 and LAIR-2 plasma levels differ between the acute and chronic phase after myocardial infarction in patients. Flow cytometry showed that LAIR-1 receptor expression on monocytes (a) 3 days after MI was higher than 6 weeks thereafter and compared to healthy controls, which could be mainly attributed to CD14 ++ CD16 − classical (b) and CD14 ++ CD16 + intermediate monocytes (c), but not to CD14 − CD16 + non-classical monocytes (d). Similar to monocytes, also granulocytes showed higher LAIR-1 receptor expression in the acute compared to the chronic phase, but no difference between the chronic phase and healthy controls was observed (e). No difference was observed in LAIR-1 receptor expression on CD4 + T-lymphocytes (f). Though not significant, sLAIR-1 was higher 3 days after MI compared to 6 weeks (g). In contrast, plasma levels of LAIR-2 were lower 3 days after MI compared to 6 weeks on cells of myeloid origin, amongst which neutrophils, macrophages and Ly6C expressing monocytes (Fig. 5a). While LAIR-1 expression was similar over time in T-cells and monocytes, we did observe a decrease in LAIR-1 expression on macrophages 3 days and on neutrophils 28 days after MI (Supplementary Fig. S4). As expected, LAIR-1 was undetectable on cells from LAIR-1 −/− mice. Flow cytometry was performed for the characterization and quantification of leukocytes in the blood, spleen, draining lymph node and heart after MI (see Supplementary Fig. S2). Three days after MI, a robust leukocyte influx in the heart was observed, that mostly consisted of neutrophils and CD8 + T-cells (Table 2). No difference in white blood cell subtype composition was observed between WT and LAIR-1 −/− mice 3 days and 28 days after MI in all studied organs (Table 2 -blood, infarct area; see Supplementary Table S3-S4 -remote area, lymph nodes, spleen).

Discussion
Leukocytes and leukocyte activation, in particular monocytes and neutrophils, have been shown to play an important role in both cardiac ischemia-reperfusion injury and remodelling [22][23][24][25] . LAIR-1 is present on a variety of immune cells 14 and important in the regulation of leukocyte activation in response to an inflammatory reaction 15,26,27 . We observed increased LAIR-1 expression on leukocytes of patients compared to healthy controls. More specifically, LAIR-1 expression on circulating monocytes and neutrophils is increased directly after MI and declines after six weeks, suggestive of immune regulation by LAIR-1 in a response to the pro-inflammatory environment of MI. In more detail, LAIR-1 expression differed on pro-inflammatory CD14 ++ CD16 − classical and CD14 ++ CD16 + intermediate monocytes. Although both are necessary for the removal of debris following MI, their effect is generally considered disproportionate and detrimental 11 . Therefore, higher LAIR-1 expression in the acute phase after MI may be beneficial in suppressing pro-inflammatory monocyte activation to limit cardiac damage. Although LAIR-1 expression on monocytes decreases in the chronic phase after MI, the expression levels remain increased when compared to healthy controls. This is most probably linked to ongoing low-grade inflammatory response in the chronic phase of cardiac remodelling after MI 28 . In addition, we observed higher LAIR-1 expression on neutrophils in the acute phase after MI compared to the chronic phase. Considering the observation that pro-inflammatory stimuli lead to the higher LAIR-1 expression on neutrophils 29 , this is in agreement with the strong inflammatory response directly after MI. Next to increased LAIR-1 expression on monocytes and neutrophils, we also observed higher levels of sLAIR-1 in the acute phase after MI, which is in line with the observation that cell activation induces shedding of LAIR-1 30 . Although the source of sLAIR-1 remains to be elucidated, this finding suggests that inflammation in the acute setting of MI increases LAIR-1 expression even more to result in both high expression levels and a high amount of LAIR-1 shedding. Contrarily, the levels of LAIR-2, mainly produced by stimulated CD4 + T-lymphocytes, are lower in the acute phase of MI. This difference might be (partially) explained by the relatively decreased number of circulating CD4 + T-cells in the acute stage after MI compared to the chronic phase 31 . Though both sLAIR-1 and LAIR-2 are natural antagonists of cell-bound LAIR-1, LAIR-2 has been shown to be far more potent than sLAIR-1.
These findings in patients prompted us to study if LAIR-1 is causally involved in ischemia reperfusion injury in the heart. However, in mice, the absence of LAIR-1, did not affect infarct size or cardiac remodelling after MI. Although leukocyte activating receptors and inflammation are widely recognized as important players in ischemia-reperfusion injury and remodelling after MI [32][33][34] , and despite the regulation in LAIR expression in MI patients, we were not able to establish a causal role for LAIR-1 deficiency in this regard.
The extent of ischemia-reperfusion injury is in agreement with previously performed experiments in WT mice in our laboratory [34][35][36] , infarct size after myocardial ischemia-reperfusion did not differ between WT and LAIR-1 −/− mice. We anticipated on increased reperfusion injury in the LAIR-1 −/− mice as a consequence of enhanced cellular infiltration and inflammation. However, the inflammatory response assessed in various tissues in the acute (3 days) and more chronic (28 days) inflammatory phase after MI did not differ between WT and LAIR-1 −/− mice. In addition, the deposition of collagen, as well as the extent of cardiac remodelling at 28 days was comparable to those observed in previously performed experiments 37,38 , but did not differ between both groups.
Activating leukocyte receptors 34,39,40 and costimulatory molecules 41 have been shown to play an important role in myocardial reperfusion injury through modulation of the inflammatory response, whereas studies on inhibitory receptors or co-inhibitory molecules are lacking. The inhibitory effect of LAIR-1 may not provide sufficient potency for the extent of tissue damage and severity of the inflammatory response in the present model, as was previously shown in vitro 42 . This is in agreement with the observation that LAIR-1 has been shown to be primarily involved in low-grade chronic inflammatory diseases, such as cancer 19,43 and chronic contact dermatitis 44 , but not so much in acute, high grade, inflammatory responses as observed in experimental autoimmune encephalitis and LPS injection 45 . Although the chronic phase of myocardial remodelling shows a somewhat less inflammatory response than the acute phase of experimental MI, leukocyte influx is still impressive 46 .
In addition, cell activation starts in the bloodstream 47 , whereas inhibition of LAIR-1 is expected to occur predominantly upon the encounter of collagen in the heart. This may either be too late to efficiently inhibit the already initiated pro-inflammatory cascade, or the sheer amount of collagen ligands is too low to induce robust activation of LAIR-1. Moreover, other inhibitory receptors and/or pathways could have compensated for the absence of LAIR-1.
In conclusion, LAIR-1 expression on monocytes and neutrophils is increased in patients 3 days after MI. Though, in mice, the absence of LAIR-1 does not influence infarct size, nor does it affect inflammation, fibrosis formation and adverse left ventricular remodelling in mice four weeks after acute MI.