Cellular & Molecular Immunology (2014), 1–12 ß 2014 CSI and USTC. All rights reserved 1672-7681/14 $32.00 www.nature.com/cmi

RESEARCH ARTICLE

Stretch-induced human myometrial cytokines enhance immune cell recruitment via endothelial activation Yu-Hui Lee1,2, Oksana Shynlova1,3 and Stephen J Lye1,2,3 Spontaneous term labour is associated with amplified inflammatory events in the myometrium including cytokine production and leukocyte infiltration; however, potential mechanisms regulating such events are not fully understood. We hypothesized that mechanical stretch of the uterine wall by the growing fetus facilitates peripheral leukocyte extravasation into the term myometrium through the release of various cytokines by uterine myocytes. Human myometrial cells (hTERT-HM) were subjected to static mechanical stretch; stretch-conditioned media was collected and analysed using 48-plex Luminex assay and ELISA. Effect of stretch-conditioned media on cell adhesion molecule expression of human uterine microvascular endothelial cells (UtMVEC-Myo) was detected by quantitative polymerase chain reaction (qPCR) and flow cytometry; functional assays testing leukocyte–endothelial interactions: adhesion of leukocytes to endothelial cells and transendothelial migration of calcein-labelled primary human neutrophils as well as migration of THP-1 monocytic cells were assessed by fluorometry. The current in vitro study demonstrated that mechanical stretch (i) directly induces secretion of multiple cytokines and chemokines by hTERT-HM cells (IL-6, CXCL8, CXCL1, migration inhibitory factor (MIF), VEGF, G-CSF, IL-12p70, bFGF and platelet-derived growth factor subunit B (PDGF-bb), P,0.05); stretch-induced cytokines (ii) enhance leukocyte adhesion to the endothelium of the surrounding uterine microvasculature by (iii) inducing the expression of endothelial cell adhesion molecules and (iv) directing the transendothelial migration of peripheral leukocytes. (vi) Chemokine-neutralizing antibodies and broad-spectrum chemokine inhibitor block leukocyte migration. Our data provide a proof of mechanical regulation for leukocyte recruitment from the uterine blood vessels to the myometrium, suggesting a putative mechanism for the leukocyte infiltrate into the uterus during labour and postpartum involution. Cellular & Molecular Immunology advance online publication, 2 June 2014; doi:10.1038/cmi.2014.39 Keywords: endothelium; labour; leukocyte infiltration; stretch; uterus

INTRODUCTION A successful pregnancy and delivery requires the adaptation of various uterine tissues to the growing fetus. During the most part of pregnancy, an intact cervix and fetal membranes are necessary to retain the conceptus in utero, while the uterine muscle (myometrium) undergoes several stages of transformation to accommodate the developing fetus, placenta and amniotic fluid. Near term, extensive remodelling occurs at pregnancy-associated sites to prepare for delivery: rupturing of the fetal membrane, softening and effacement of the cervix for dilation and activation of the myometrium to coordinate synchronous contractions to expel the fetus. Multiple research groups have suggested the involvement of the maternal immune system in the regulation of these events during term labour (TL). 1

It has been proposed that spontaneous human TL is a culmination of interlinked physiological processes that are dominated by the inflammatory system to switch from a quiescent state of the uterus during pregnancy to an activated state during labour contractions.1–3 TL now is recognized as a localized inflammatory reaction characterized by (i) the activation of maternal peripheral leukocytes;4–7 (ii) the increased expression of cytokines/chemokines by gestational tissue;8–10 and (iii) upregulation of cell adhesion molecules (CAMs) on uterine vascular endothelial cells.11 We recently proposed a new model of TL as a physiological uterine inflammation process involving multiple uterine compartments, i.e., the myometrium, decidua and cervix that lead to the successful delivery of baby.7 It was reported that moderate to marked inflammation indicated by the presence of leukocytes

Lunenfeld-Tanenbaum Research Institute, Mount Sinai Hospital, Toronto, Canada; 2Department of Physiology, University of Toronto, Toronto, Canada and Department of Obstetrics & Gynecology, University of Toronto, Toronto, Canada Correspondence: Dr O Shynlova, Lunenfeld-Tanenbaum Research Institute at Mount Sinai Hospital, 25 Orde Street, Suite 6-1019, Toronto, Ont. M5G 1X5, Canada E-mail: [email protected] Received: 2 January 2014; Revised: 16 April 2014; Accepted: 5 May 2014 3

Mechanical stretch of myometrium recruits leukocytes YH Lee et al

2

in myometrial samples was detected in women after the onset of labour compared to before the onset, which correlated well with cervical dilatation and fetal membrane rupture.12 Additionally, macrophage abundance in the human decidua was higher in TL than term non-labouring samples.13 In parallel, upregulation of pro-inflammatory cytokines mRNA expression of Interleukin-1 b (IL-1b, IL-6, C-X-C motif ligand (CXCL) 8 and C-C motif ligand (CCL) 2 was observed in the myometrium of healthy labouring women.14–18 Our recent data confirmed these findings and revealed a more extensive list of up-regulated cytokine and chemokine proteins (CXCL8, CXCL1, CCL2, CCL7, IL-9, IL-18, IL-1RN, tumor necrosis factor alpha (TNFa) and granulocyte colony-stimulating factor (G-CSF)) in healthy TL human myometrial tissue samples, while also detecting a higher number of macrophage and neutrophil infiltrate.7,19 The trigger that elevates cytokines in the myometrium of healthy women during labour remains largely elusive. Mechanical stretch has been widely demonstrated as an inducer of cytokine synthesis in different tissues.20,21 In particular, many have reported the ability of various uterine cell types such as fetal membrane epithelial cells, uterine myocytes and cervical fibroblasts to secrete cytokines following mechanical stretch stimulus.22Using murine models of gestation, we have demonstrated that rat and mouse myometrial smooth muscle cells (SMCs) are capable of producing multiple cytokines under the influence of uterine stretch imposed by the growing fetus around the time of labour.23,24 With a unilateral pregnant rat model, we proved that gravidity is associated with the increased cytokine production (IL-1a, CCL2, CXCL1 and IL-6) in the myometrium during TL.7,24 Others have also shown the effect of mechanical stretch on CXCL8 synthesis in human myometrial cells and fetal membranes.8,10 We hypothesized that multiple cytokines induced by static stretch of SMCs in term pregnant human myometrium may activate the expression of CAMs on endothelial cells in uterine vasculature, which then stimulate the infiltration of peripheral leukocyte in preparation for labour. In addition, the infiltrated leukocyte may contribute to the increase in cytokine levels in the myometrium since they are rich sources of these inflammatory molecules.25 Therefore, the aim of this study is to gain direct evidence that mechanical stretch of the myometrial SMCs facilitates peripheral leukocyte transendothelial migration into the myometrium. The three main objectives of this study include: (i) the characterization of cytokines secreted in vitro by human myometrial SMCs upon artificial static mechanical stretch; (ii) the analysis of activation by these stretch-induced cytokines of primary human leukocytes (neutrophils and monocytes) and human microvascular endothelial cells (via the upregulation of CAMs); and (iii) the effect of multiple cytokines secreted by stretched myometrial cells on transendothelial migration of immune cells as an in vitro model of the peripheral leukocyte recruitment into the uterine smooth muscle. MATERIALS AND METHODS Cell culture Human myometrial smooth muscle cell line immortalized with human telomerase reverse transcriptase (hTERT-HM, a gift from Dr Jennifer Condon26) was cultured in phenol red-free DMEM/ Cellular & Molecular Immunology

F12 supplemented with 10% FBS (Gibco, Burlington, ON, Canada), 15 mM HEPES, 100 U/ml penicillin/streptomycin and 0.25 mg/ml amphotericin B (Lonza Walkersville, Inc., Walkersville, MD, USA). Human uterine microvascular endothelial cell line (hUtMVEC-Myo) was purchased from Lonza and grown in endothelial growth media (EGM-2MV; Lonza) supplemented with supplier-recommended concentrations of human EGF, hydrocortisone, gentamicin, VEGF, human bFGF, IGF, ascorbic acid and 5% FBS. Both cell lines were cultured and used from passages 4–8 in a 37uC incubator with 5% CO2. Human monocytic cell line (THP-1) was purchased from ATCC (ATCC, Manassas, VA, USA) and maintained in RPMI-1640 (Gibco) supplemented with 10% FBS, 100 U/ml penicillin/streptomycin and 0.25 mg/ml amphotericin B and 0.05 mM of 2-mercaptoethanol (Sigma-Aldrich, St. Louis, MO, USA). Culture was passaged when cell density reached 1.03106 cells/ml. Human peripheral blood collection and neutrophil isolation The study design was approved by the Institutional Review Board of Mount Sinai Hospital, Toronto. Healthy pregnant women with a singleton pregnancy presenting for regular obstetric care at the hospital were recruited and written consent to participate in the study was obtained from each patient. Peripheral blood were collected in Gel and Lithium Heparin vacutainers (BD Diagnostics, Franklin Lakes, NJ, USA) and processed within 1 h of collection. Primary human neutrophils were isolated from heparinized whole blood using a double gradient consisting of HISTOPAQUE-1119 and -1077 (Sigma-Aldrich) centrifuged at 700g for 30 min at room temperature, 25uC. Neutrophil interphase was extracted, washed and adjusted to concentration of 1.53106 cells/ml in serum-free DMEM (SF-DMEM) supplemented with ITS-A (Insulin, Transferrin, Selenium, Sodium Pyruvate Solution; Life Technologies, Grand Island, NY, USA). Isolated neutrophils were .90% pure and .95% viable as determined later by flow cytometry and trypan blue, respectively. Application of static stretch The impact of mechanical stretch on myometrial cytokine secretion was investigated in vitro using hTERT-HM cells and a vacuum-driven Flexcell computer system (FX-5000; Flexcell International Corp., Hillsborough, NC, USA). Stretch experiment was performed as previously described.24 hTERT-HM cells were plated at a density of 300 000 cells/well; when confluent, cells were serum-starved overnight prior to stretch. Static stretch up to a maximum of 25% was applied for 24 h inside a humidified incubator with 5% CO2 at 37uC. Conditioned media, both from control non-stretched (NS-CM) and stretched plates (SCM) were collected, centrifuged (10 min at 1000g, 4uC) and filtered through 0.22 mm Millex PEs syringe filter units (EMD Millipore Corp., Billerica, MA, USA) to remove cellular debris. Processed conditioned media was kept in 220uC until analysis. Characterization of stretch-induced myometrial cytokines Luminex assay. A total of 48 human cytokines were screened in the collected conditioned media using the 27-plex and 21-plex Panels of the Bio-Plex Pro Human Cytokine Assays (Bio-Rad

Mechanical stretch of myometrium recruits leukocytes YH Lee et al

3

Table 1 Real-time PCR primer sequences of a panel of genes studied and housekeeping genes Gene

Refseq mRNA accession no.

Primer sequences (59–39)

TBP

NM_003194

YWHAZ

NM_003406

SDHA

NM_004168

PECAM1

NM_000442

ICAM1

NM_000201

SELE

NM_000450

VCAM1

NM_001078

Forward: TGC ACA GGA GCC AAG AGT GAA Reverse: CAC ATC ACA GCT CCC CAC CA Forward: ACT TTT GGT ACA TTG TGG CTT CAA Reverse: CCG CCA GGA CAA ACC AGT AT Forward: TGG GAA CAA GAG GGC ATC TG Reverse: CCA CCA CTG CAT CAA ATT CAT G Forward: TCC ACC AGC GTC ATT GGC GT Reverse: TGC CCT TGC GGT GTT AGG CA Forward: GAC CGC AGA GGA CGA GGG CA Reverse: TTG GGC GCC GGA AAG CTG TAG Forward: TCT GCT GCT GGA CTC TCC CTC C Reverse: GCA GCT CTG GCA GGA ACA AA Forward: TCC AGG TGG AGC TCT ACT CAT TCC Reverse: CGG TCA AGG GGG TAC ACG CT

Laboratories Inc., Hercules, CA, USA) following manufacturer’s instruction manual. The assay plate was read on the Bio-Plex 200 System with High-Throughput Fluidics. Standards and samples were run in duplicates and analysed using the Bio-Plex Manager 5.0 to concentrations in pg/ml. During each run, SF-DMEM was assayed as blank control. ELISA. ELISA was performed to validate the multiplex assay screening results as per manufacturer’s instruction. Samples were diluted with SF-DMEM accordingly to ensure the absorbance readings stayed within the linear range of the standard curve. Human IL-6 and IL-8 (CXCL8) Ready-SET-Go! and vascular endothelial growth factor (VEGF)-A Platinum ELISA kits were purchased from eBioscience (San Diego, CA, USA). Human CXCL1 Single-Analyte ELISAarray Kit was purchased from SABiosciences (Mississauga, ON, Canada) whereas human GCSF ELISA kit was from RayBiotech Inc. (Norcross, GA, USA). Endothelial CAM expression Experimental plates were set up first by precoating 12-well culture plates for 1 h with 0.1% gelatin in sterile water (Millipore, Billerica, MA, USA) at 37uC. Human UtMVEC-Myo cells were plated onto the coated 12-well plates with a seeding density of 100 000 cells per 1.5 ml of EGM-2MV and grown to confluence (48 h). Confluent monolayers were serum-starved overnight with endothelial basal medium supplemented with ITS-A. The next day, 500 ml of SF-DMEM, NS-CM, S-CM or selected individual cytokines was added to the monolayer and incubated for 4 h for quantitative polymerase chain reaction (qPCR) analysis or for 6 h for flow cytometry analysis. Real-time qPCR analysis We investigated the effect of S-CM on the gene expression of Eselectin, VCAM-1, ICAM-1 and PECAM-1 normalized against three housekeeping genes (YWHAZ, TBP and SDHA). Total RNA was extracted from UtMVEC-Myo cells and purified using the RNeasy Mini Elute Kit (Qiagen, Germantown, MD, USA) with the use of genomic DNA Eliminator columns. RNA concentration was measured by the NanoDrop 1000 spectrophotometer

(Thermo Fisher Scientific Inc., Wilmington, DE, USA). RNA quality was assessed using the Experion RNA StdSens Analysis Kit (Bio-Rad Laboratories Inc.) following manufacturer’s instructions. Samples with a RQI value greater than 7.5 and a clean separated gel image were accepted as intact RNA. cDNA were generated with iScript reverse transcription supermix (Bio-Rad Laboratories Inc.). All PCR reactions were carried out in triplicates on the CFX384 Real Time System C1000 Thermal Cycler (Bio-Rad Laboratories Inc.) with 5 ng of cDNA, LuminoCt SYBR Green Qpcr ReadyMix (Sigma-Aldrich), forward and reverse primers at a final concentration of 300 nM (Table 1). Gene expression values were analysed using the DDCq mode on the CFX Manager software 3.0 (Bio-Rad Laboratories Inc.). Flow cytometry Fluorochrome-conjugated mouse anti-human antibodies were purchased from BD Pharmingen, Mississauga, ON, Canada: CD54-APC, CD62E-PE-Cy 5, CD106-PE, CD11b-PE-Cy 5, CD44-PE, CD45-APC-H7, CD14-PerCP and CD15-FITC. The protein expression of E-selectin (CD62E), VCAM-1 (CD106) and ICAM-1 (CD54) was analysed on the surface of stimulated hUtMVEC-Myo cells. Experimental protocol was modified from Gra¨bner et al.27 Briefly, stimulated monolayers were washed and stained with LIVE/DEAD Fixable Violet Stain for 20 min in the dark (LDVio; Molecular Probes Inc., Eugene, OR, USA). hUtMVEC-Myomonolayer was fixed, stained for specific CAMs for 30 min at 4uC and detached with trypsin/ EDTA for 3 min at 37uC. Stained hUtMVEC-Myo cells were collected, washed (400g, 4uC, 6 min) and resuspended with BD Stabilizing Fixative (BD Biosciences). To study the protein expressions of CD44 and CD11b on monocytes and neutrophils, we collected peripheral blood from healthy first-trimester pregnant women. Whole blood was stimulated for 1 h (CD11b) or 6 h (CD44) with NS-CM, S-CM or SF-DMEM (1 : 10 ratio) followed by FACS analysis. Basal expression of surface antigens was established by staining the time ‘zero’ SF-DMEM tube. Following stimulation, whole blood mixtures were stained for dead cells (20 min, 4uC), CD44 and CD11b (30 min, 4uC). Stained samples were transferred Cellular & Molecular Immunology

Mechanical stretch of myometrium recruits leukocytes YH Lee et al

4

into Fix/Lyse Solution (BD Biosciences) and kept at 4uC until analysis. Each data point was normalized to the basal CD11b or CD44 expression at time 0. Stimulated leukocytes and hUtMVEC-Myo were gated first with negative LDVio staining against forward scatter. Leukocytes were gated by forward-side scatter plots and subsequently by positive CD45 staining; granulocytes were identified as CD451CD151and monocytes as CD451CD141. For both leukocyte and EC populations, doublet discrimination was achieved by gating forward and side scatter against their corresponding time of flight. Fluorescence minus one control (FMO) tubes were prepared for every treatment to set appropriate gates during analysis. Data acquisition was run using the 10-colour Gallios Flow Cytometer (Beckman Coulter, Inc., Mississauga, ON, Canada) with at least 5000 hUtMVEC-Myo cells or 1000 monocytes as a stop mark. Data analysis was performed using KaluzaAnalysis Software v.1.2 (Beckman Coulter, Inc., Mississauga, ON, Canada). In each experiment, compensation beads (BD Biosciences) stained with each antibody were prepared and run with the samples. To standardize fluorescent readings of individual experimental runs, the instrument was calibrated each time using Flow-Set Pro Fluorospheres (Beckman Coulter, Inc.) prior to data acquisition. Permeability assay BD Falcon 3.0-mm cell culture inserts for 24-well plate (BD Biosciences) were precoated with 0.1% gelatin for 1 h at 37uC. hUtMVEC-Myo cells were grown in the insert (75 000 cells per 500 ml) to form a monolayer for 72 h. Monolayer confluence was tested measuring its permeability to trypan blue/BSA (adapted from Fiuza et al.28). Inserts were washed and transferred into clean wells containing 1 ml HBSS. Confluent monolayers on the insert were serum-starved overnight with SF-DMEM. The next day, hUtMVEC-Myo cells were stimulated with NS-CM or S-CM on both sides of the insert for 4 h at 37uC. Permeability of the monolayer after stimulation was examined by incubation with 1 mg/ml FITC-dextran in HBSS (average molecular weight 3000–5000 g/mol; Sigma-Aldrich) for 1.5 h. Two 150 ml aliquots of the bottom well solution were transferred to a 96-well black reading plate for fluorescence measurement at 490/520 nm. A 10-point standard curve was generated from a 1 : 2 dilution series with 1 mg/ml FITC-dextran as the starting concentration. A non-coated insert without hUtMVEC-Myo cells was used as a control for maximum leakage in each experiment. Additionally, two endothelium-coated inserts with confluent monolayers were incubated with endothelial growth media only (no stimulation) during the 4-h stimulation period and used as a negative control for minimal leakage of an intact monolayer. Adhesion assay Protocol was adapted and modified from Fiuza et al.28 Human UtMVEC-Myo cells were grown in 96-well culture plates (10 000 cells/well) precoated with 0.1% gelatin to form a monolayer for 48 h. After overnight serum starvation, the monolayer was incubated with NS-CM or S-CM for 3 h. Neutrophils were labelled with 2.5 mMcalcein-AM (eBiosciences) for 30 min. Cellular & Molecular Immunology

After endothelial cell stimulation, calcein-labelled human neutrophils (100 000 neutrophils in 150 ml of SF-DMEM per well) were added and allowed to adhere for 1 h at 37uC. Nonadherent cells were removed by washing with warm PBS. Fluorescence of the remaining adherent cells was quantified on Infinite 200 (Tecan Group Ltd., Mannedorf, Switzerland) at 490/520 nm. Number of adherent cells was calculated against a standard curve generated from 1 : 2 serial dilutions of calceinAM-labelled neutrophils (from 200 000 to 3125). Transendothelial migration (TEM) and chemotaxis assays Endothelial cell-coated inserts were performed as previously described. Confluent inserts were stimulated with NS-CM/SCM or selected individual chemokines (CXCL8, CXCL1; R&D Systems, Inc., Minneapolis, MN, USA) at varying concentrations for 3–4 h. Primary human neutrophils were labelled with 2.5 mM calcein-AM (eBiosciences) for 30 min, loaded to the 3.0-mm inserts (2.03105 cells in 150 ml of SF-DMEM) and transmigrated for 1 h. In some experiments, neutralizing antibodies against CXCR1, CXCR2 or both receptors (10 mg/ml; R&D Systems, Inc.) were incubated with the isolated neutrophils 30 min prior to the TEM experiment. Transmigrated calcein-labelled human neutrophils were collected from the bottom chamber and measured for fluorescence at 490/520 nm. Monocyte (THP-1) chemotaxis towards conditioned media was tested using CytoSelect 24-Well Cell Migration Assay (Cell Biolabs, Inc., San Diego, CA, USA). THP-1 cells were labelled with 2.5 mM calcein-AM for 30 min, loaded into the 8.0-mm inserts (2.03105 cells in 100 ml of SF-DMEM) and migrated for 2 h. Effect of broad-spectrum chemokine inhibitor (BSCI) inhibitor The inhibitory effect of BSCI (Funxional Therapeutics, Cambridge, UK) on primary human neutrophil TEM and THP-1 chemotaxis towards S-CM was examined to assess the potential effect of chemokine blockage on leukocyte recruitment. For TEM, S-CM was pretreated with 2 nM of BSCI in DMSO for 1 h at 37uC prior to endothelial stimulation, for THP-1 migration S-CM was pretreated with 5 nM of BSCI, or with vehicle (DMSO) as a negative control. Statistical analysis Data normality was assessed by the Kolmogorov–Smirnov test. For analysis of two groups, unpaired and paired t-tests were done for normally distributed data and Mann–Whitney U test for nonparametric data. For comparison between more than two groups, one-way ANOVA was employed. Post-tests such as Bonferroni or Tukey’s test were used and specified in each specific analysis section. Significance level was set at *P,0.05, **P,0.01 and ***P,0.001. All statistical analyses were done using Prism 5 (GraphPad Software, Inc., La Jolla, CA, USA). RESULTS Human myometrial cytokine expression induced by static mechanical stretch Conditioned media from non-stretched and stretched hTERTHM cell cultures were collected after undergoing 24 h of 25%

Mechanical stretch of myometrium recruits leukocytes YH Lee et al

5

static stretch. A total of 48 cytokines were analysed and the expression of nine cytokines was found significantly increased by stretch (Figure 1a). These include IL-6, IL-12p70, CXCL8, CXCL1, macrophage migration inhibitory factor (MIF), GCSF, basic fibroblast growth factor (bFGF), VEGF and platelet-derived growth factor subunit B (PDGF-bb). These stretch-induced cytokines are essential players during leukocyte extravasation that function as either pro-inflammatory agents (IL-12p70, MIF, IL-6, CXCL1, CXCL8) or growth factors (G-CSF) affecting leukocytes and endothelium (VEGF). Among these cytokines, CXCL8 and CXCL1 had the highest stretch-induced increase in S-CM as compared to NS-CM controls (Figure 1b, fold changes of 15.7 and 16.1, respectively).

Concentration (pg/ml) [log]

a

10000

NS-CM S-CM

*** ***

**

1000

*

* *

100

*

*

10

*

VE G F

C SF 3 IL 12 p7 0

VE G F

C SF 3

bF G F PD G FB

M IF

L1 XC

L8

M IF

b

C

XC C

IL 6

1

25

Relative Fold Change

20

15

10

5

PD G FB

p7 0 bF G F

12 IL

L1 XC C

L8 XC C

IL 6

0

Figure 1 Pro-inflammatory cytokines are up-regulated by static stretch in hTERT-HM cells. Cytokines released by hTERT-HM cells in culture media upon 24 h of 25% static stretch (n57). White bars represent cytokines detected in the non-stretched conditioned media (NS-CM), whereas black bars represent cytokines detected in the stretched conditioned media (S-CM). (a) Absolute concentration values are presented as mean6s.e.m. on a logarithmic scale. (b) Increase in cytokine levels secreted in S-CM is illustrated as fold change relative to NS-CM levels (the dotted line in the bottom designates 1). Significance was set at *P,0.05 comparing each S-CM group to its corresponding NS-CM group.

Activation of uterine microvascular endothelial cells by stretch-induced cytokines To initiate leukocyte extravasation, the expression of CAMs on the endothelium neighbouring the inflammation site must be increased.29 Therefore, we investigated the effect of S-CM on the expression of four CAM genes in human myometrial microvascular endothelial cells. We observed that mRNA levels of E-selectin, VCAM-1 and ICAM-1 were significantly upregulated by S-CM treatment when compared to NS-CM (17.2-, 8.5- and 3.6-fold increase, respectively), whereas PECAM-1 expression was unaffected (Figure 2a). Next we examined the protein expression of three CAMs whose gene expression was found to be upregulated by S-CM (E-selectin, VCAM-1 and ICAM-1) via flow cytometry analysis. The distribution of these three CAMs on hUtMVEC-Myo cells is demonstrated by Figure 2c, where ICAM-1 protein is expressed constitutively; however, E-selectin and VCAM-1 show low basal protein levels. After treatment with NS-CM and S-CM, ICAM-1 expression on endothelial cells was progressively increased in comparison to negative control (see right shift on histogram plot vs. SF-DMEM), with S-CM being the most potent stimulator (Figure 2b). The MFI value, indicative of surface ICAM-1 protein expression, increased significantly (P,0.05) from 49.9068.9units (NS-CM) to 65.80613.83 units (S-CM) on a logic scale (Figure 2d).E-selectin protein expression was detected on 2.23%61.18% of hUtMVEC-Myo cells after incubation with NS-CM. S-CM stimulation further increased E-selectin-positive endothelial cells by 2.8-fold to 6.25%61.91%. VCAM-1 expression was detected on 3.38%61.53% of hUtMVEC-Myo cells following NS-CM treatment, whereas S-CM treatment increased the expression by 2.60-fold to 8.78%63.24%. These results suggest the potential role of stretch-induced cytokines in the activation of myometrial microvascular endothelial cells. Further, we explored whether S-CM also influences vascular permeability in parallel with activating endothelial cells. In vitro analysis of FITC-dextran dye leakage through the endothelial monolayer treated with S-CM revealed significant increase of the permeability compared to the negative control (P50.0248; Figure 2e). No difference in leakage, however, was observed between cells treated with NS-CM and S-CM, which indicate that multiple cytokines secreted by myometrial SMCs likely were able to influence the integrity of the uterine vasculature in the absence of stretch. Peripheral leukocyte activation by stretch-induced myometrial cytokines Next, using flow cytometry analysis of CD11b and CD44 protein expression, we explored the activation status of peripheral leukocytes following stimulation with NS-CM and S-CM. Peripheral blood samples from pregnant women were analysed for the expression of CD11b after 1-h incubation with CM. Our results showed that treatment with NS-CM and S-CM elicited significant CD11b activation on the surfaces of both monocytes and neutrophils when compared to treatment with negative control SF-DMEM (Figure 3a). However, no difference was observed between NS-CM and S-CM stimulation. These data Cellular & Molecular Immunology

Mechanical stretch of myometrium recruits leukocytes YH Lee et al

6

a

30

b

**

FMO

S-CM 20

SF-DMEM hUtMVEC-Myo Count

Relative mRNA Expression (Fold Change)

60

** 10

***

NS-CM S-CM ICAM-1

0 0

101

100

102

103

0 ICAM1

PECAM1

d

100 NS-CM S-CM

80 60 40

*

20

*

0

CAM Protein Expression (MFI)

CAM Positive (% of total live hUtMVEC-Myo)

c

VCAM1

ICAM-1

e

100

*

100

80 60 40 20 0

ICAM-1

E-selectin

VCAM-1

Endothelial Permeability (FITC-dextran leakage of uncoated control, %)

SELE

ICAM-1

E-selectin

VCAM-1

80

* 60 40 20 0 Control

NS-CM

S-CM

Figure 2 Stretch-conditioned media activates human microvascular endothelial cells through enhanced cell adhesion molecule expression and vascular permeability. (a) Confluent endothelial cells (hUtMVEC-Myo) were stimulated with NS-CM or S-CM for 4 h and the expression of E-selectin, ICAM-1, VCAM-1 and PECAM-1 genes were analysed by RT-PCR (unpaired t-test, n53). Expression of cell adhesion molecules on S-CM-treated hUtMVEC-Myo is presented as fold change relative to corresponding expression of NS-CM-treated cells (the dotted line in the bottom designates 1). (b) Representative FACS histogram overlay comparing surface protein expression of ICAM-1 on hUtMVEC-Myo (n56) after incubation with serumfree media (SF-DMEM, blue), NS-CM (green) and S-CM (red). FMO controls (grey) were used to determine the boundary of individual gates. (c) Gated values (percentage of positive cells) and (d) MFI values of cell adhesion molecule proteins expressed on hUtMVEC-Myo treated with NS-CM (white bars) and S-CM (black bars). Data was analysed by unpaired t-test (n56). (e) Confluent hUtMVEC-Myo grown on 3.0-mm inserts were stimulated with NS-CM (white bars), S-CM (black bars) or endothelial growth media (control, striped bars) for 4 h and the leakage of FITC-dextran was analysed by fluorescence. Bar graphs represent leakage percentage of each treatment relative to the maximum leakage (100%) detected in the absence of endothelial cells. Data was analysed by one-way ANOVA with Tukey’s post-test (n54). For a, c, d and e: *P,0.05, **P,0.01, ***P,0.001 compared to NS-CM group; #P,0.05 compared to control. All values are presented as mean6s.e.m. FMO, fluorescence minus one; MFI, median fluorescence intensity.

indicate that multiple cytokines secreted by myometrial cells, irrespective of mechanical stretch, were able to activate the expression or the translocation of the integrin molecule to the surface of white blood cells. On the contrary, CD44 expression on both peripheral monocytes (1.24-fold) and neutrophils (1.22-fold) was increased after 6-h incubation with S-CM when compared to NS-CM treatment (Figure 3b). Stretched-induced myometrial cytokines enhanced leukocyte recruitment To mimic the effect of mechanical stretch on the adhesion of immune cells to the nearby endothelium in the myometrium, we stimulated hUtMVEC-Myo monolayer with media conditioned by myometrial cells and examined the ability of primary human neutrophils to adhere. Our data indicate that S-CM Cellular & Molecular Immunology

significantly increased the number of bound neutrophils compared to NS-CM (1.69-fold, P50.0286; Figure 4a). We next asked whether the stretched-induced secretion of multiple chemokines is capable of stimulating chemotaxis of immune cells. TEM assay was performed to model neutrophil extravasation into the myometrial tissue in the presence of mechanical stretch. We found that S-CM treatment significantly induced neutrophil transmigration as compared to NSCM treatment (1.72-fold increase of immune cells detected in the bottom chamber, P50.0096, Figure 4b). Two chemokines most significantly upregulated by stretch in the hTERT-HM conditioned media (CXCL1 and CXCL8) are neutrophil chemoattractants exerting their activities through CXC chemokine receptors 1 and 2. Thus, we incubated peripheral neutrophils with anti-CXCR1 and anti-CXCR2 neutralizing antibodies

Mechanical stretch of myometrium recruits leukocytes YH Lee et al

7

a FMO BASAL

15 Monos Count

Grans Count

200

SF-DMEM NS-CM S-CM

100

#1

#1 0

0 101

0

102

101

0

103

102

103

CD11b

CD11b

b 15

Grans Count

Monos Count

300 10

5 #1

#1 0

0 0

100

101

102

103

CD44

0

100

101

102

103

CD44

Figure 3 Multiple cytokines released from human myometrial smooth muscle cells activate peripheral leukocytes. Representative FACS histogram overlay plots of CD11b and CD44 protein expression on surface of peripheral granulocytes (Grans) and monocytes (Monos) following stimulation with SF-DMEM (blue, negative control), NS-CM (green) and S-CM (red). Fluorescence minus one controls (‘FMO’, grey) were used to determine the boundary of individual gates. Basal expression levels of both leukocyte protein markers prior to activation (at time ‘0’) are represented by ‘BASAL’ (magenta). (a) CD11b expression was examined after 1 h of stimulation (representative of five separate experiments). (b) CD44 expression was examined after 6 h of stimulation (representative of three experiments).

prior to TEM assay and found that TEM was reduced by 61.1% and 62.3%, respectively (P,0.05; Figure 4b) demonstrating that increased migratory characteristics of primary human neutrophils involves activation of those CXC receptors. Simultaneous antibody neutralization of both receptors further decreased TEM by 80.6% (Figure 4b). These results indicate that the enhancement of TEM by S-CM was caused, at least in part, by biologically active CXCL1 and CXCL8 secreted by myometrial SMCs after mechanical stimulation. To mimic the effect of S-CM, we tested CXCL1 and CXCL8 at various concentrations for their ability to trigger TEM of primary human neutrophils. Neutrophils from pregnant women exhibited dose-dependent increase in TEM ability (from 1.24-fold to 6.62-fold relative to control) in response to increasing concentrations of CXCL8 (from 1 to 100 ng/ml; Figure 4c). Chemotactic activity of CXCL8 was blocked by neutralizing CXCR1, whereas neutralization of CXCR2 showed no effect on percentage of transmigrated neutrophils. This likely suggests that CXCL8 mainly exerted chemotactic signalling via CXCR1 but not CXCR2. CXCL1 also enhanced neutrophil TEM in a dose-dependent manner (Figure 4d). Chemotactic activity of CXCL1 was blocked by neutralizing CXCR2 (P50.032), indicating that CXCL1-induced neutrophil TEM was mediated by CXCR2. Additionally, it was reported that BSCI simultaneously

blocks the actions of multiple chemokines in directing leukocyte chemotaxis.4,5,30 Adding BSCI to S-CM 1 h prior to the TEM assay reduced neutrophil TEM by 70.4% (P,0.05) in comparison to the S-CM group without BSCI, a level similar to the NS-CM group (Figure 4e). We next investigated whether the increased secretion of multiple myometrial cytokines in response to stretch influences chemotaxis of monocytes. Migration of monocytic THP-1 cells towards S-CM was greater than towards NS-CM (4.93-fold to 3.92-fold relative to SF-DMEM control; Figure 4f). Pretreatment of S-CM with BSCI prior to the chemotaxis assay reduced THP1 migration to a level similar to the NS-CM group (Figure 4f). These results suggest the potential inhibitory role of BSCI in leukocyte extravasation. DISCUSSION Only a few studies to date have investigated the effect of artificial mechanical stretch on the secretion of cytokines by human myometrial SMCs (mimicking biological mechanical stretch of the uterine wall by the growing fetus at late gestation) with Hua et al.31 reporting that mechanical stretch augmented the expression of five from six chemokines they analysed in primary human myometrial myocytes.31This current paper provides a detailed analysis of 48 cytokines secreted in vitro by human Cellular & Molecular Immunology

Mechanical stretch of myometrium recruits leukocytes YH Lee et al

8

b Relative Neutrophil TEM

1.0 0.5

10

100

100

+

-

+

Anti-CXCR1: -

-

-

-

+

-

Anti-CXCR2: -

-

-

+

+

Anti-CXCR2: -

-

-

-

-

+

e

-

-

-

-

+

C N

S-

nM

Anti-CXCR2: -

I5

-

C

+

CXCL1 (ng/ml)

BS

-

Relative THP-1 Migration

0 M

-

0

2

SC

-

1

***

4

SM

Anti-CXCR1: -

2

M

100

3

***

6

N

300

10

*

4

nM

0

f

I2

Relative Neutrophil TEM

5

300

100

CXCL8 (ng/ml)

-

*

300

1

Anti-CXCR1: -

10

Control

Control

S-CM

SC

***

15

2

S-

C

M

+B

d

S-CM

M

NS-CM

4

NS-CM

C

Control

6

0

0.0

0

Relative Neutrophil TEM

1.5

+

1

**

2.0

**

M

2

***

8

*

SC

3

c 2.5

S-

Relative Neutrophil Adherence

*

Relative Neutrophil TEM

a

Figure 4 Stretch-conditioned media induces adhesion of peripheral human leukocytes to endothelial cells and transendothelial migration. (a) Confluent hUtMVEC-Myo cells grown on cell culture plate were stimulated for 3 h with conditioned media. Calcein-labelled primary human neutrophils adhered to stimulated endothelial monolayer (NS-CM, white bar; S-CM, black bar). Bar graphs represent fold changes of adherent neutrophil number relative to serum-free media control. Data were analysed by unpaired t-test (n54). (b–d) Peripheral human neutrophils pretreated with either anti-CXCR1, anti-CXCR2 or both blocking antibodies (10 mg/ml) were labelled with calcein-AM and incubated with (a) conditioned media-stimulated (NS-CM or S-CM), (c) CXCL8-stimulated or (d) CXCL1-stimulated endothelial cells grown on membrane inserts (grey bars). Primary neutrophil TEM (percentage of transmigrated cells) was measured. Serum-free media (SF-DMEM) served as controls in c and d (striped bars). Graphs are representative of two independent experiments with three replicates in each treatment group. (e) In addition, BSCI (2 nM, checkered bar) was added to the S-CM and TEM was measured (n54). (f) Monocytic cell line THP-1 migration towards NS-CM/S-CM and BSCItreated S-CM was also examined (n55). (e and f) Results are shown as fold change of migrated neutrophils or THP-1 cells through membrane inserts relative to SF-DMEM control. (b–f) Values are presented as mean6s.e.m. Data was analysed by one-way ANOVA followed by Tukey’s multiple comparison test. *P,0.05, **P,0.01, ***P,0.001, compared to NS-CM group or control. BSCI, broad-spectrum chemokine inhibitor; TEM, transendothelial migration.

myometrial SMCs upon artificial mechanical stretch. Our results demonstrate that after 24 h of 25% static stretch: (i) multiple pro-inflammatory cytokines (IL-6, IL-12p70, MIF), chemokines (CXCL1, CXCL8) and growth factors (G-CSF, VEGF, bFGF and PDGF-bb) were significantly induced; (ii) these stretch-induced cytokines were able to activate human leukocytes and human microvascular endothelial cells; and (iii) increased TEM of immune cells in vitro which mimics peripheral leukocyte recruitment into the uterine smooth muscle in vivo. All cytokines induced by static stretch are known to exhibit important roles in the activation of uterine vascular endothelium and peripheral leukocytes to promote their survival, differentiation and extravasation from the blood vessels into the underlying tissues (i.e., myometrium). Importantly, our current in vitro data directly matches the cytokine profile recently detected in vivo in human labouring myometrial Cellular & Molecular Immunology

samples when compared to myometrial samples collected before labour onset, particularly in the expression of IL-6, CXCL8, CXCL1 and G-CSF.7,19 This concordance suggests that these cytokines/chemokines elevated in labouring myometrium in vivo could potentially be attributed to uterine stretch. The specific role of these cytokines in the initiation of labour is not clear. It is known that the actions of IL-6 are closely associated with labour progression. IL-6 promotes synchronous myometrial contraction via PG synthesis and expression of oxytocin and its receptor.32–35 Furthermore, IL-62/2 mice showed impaired leukocyte accumulation, suggesting that IL6 also plays a role in leukocyte recruitment.36 Two CXC family chemokines (CXCL1 and CXCL8) are well known for their strong chemoattractant activity on neutrophils, and their expression in human myometrium during spontaneous labour was reported and confirmed by many research groups.14,37,38

Mechanical stretch of myometrium recruits leukocytes YH Lee et al

9

Recent reports of transcriptome data in spontaneous human labouring myometrial samples identified IL-6, CXCL1 and CXCL8 as highly expressed genes, with CXCL8 exhibiting the highest fold change in TL compared to term not-in-labour samples.3,37 In line with our observations, Hua et al.31 showed that mechanical stretch augmented the expression of various chemokines in primary human myometrial myocytes including CCL2, CXCL8 and CXCL1. Interestingly, growth factors can also act as chemoattractants and are involved in ECM remodeling and cytokine release.39 G-CSF stimulates the proliferation and differentiation of neutrophils while promoting the mobilization of neutrophils out of bone marrow,40 suggesting that myometrial cytokine secretion may partially contribute to the neutrophilia seen in healthy pregnant women.41 In addition, G-CSF and MIF enhance neutrophil survival by inhibiting pro-apoptotic events.42 Other stretch-induced growth factors, VEGF, bFGF and PDGF-bb, have been shown to participate in the wound healing process by inducing angiogenesis,39,43 indicating their potential role in the postpartum involution of the uterus. Further, VEGF has also been shown to enhance neutrophil and monocyte TEM via inducing endothelial secretion of CXCL8 and increasing monocyte integrin expression.44 In addition, we confirmed here that stretchinduced CXCL1 and CXCL8 were biologically active since blocking either or both of CXCR1 and CXCR2 receptors on primary human neutrophils resulted in significant reduction of TEM. Furthermore, inactivating multiple chemokines simultaneously through the administration of BSCI significantly decreased both neutrophil TEM and THP-1 migration. The first step in the process of leukocyte TEM is adhesion to the activated vasculature. In particular, endothelial CAMs play an essential role in the development of a localized inflammatory response.45 It has well been known that pro-inflammatory cytokines are potent stimulator of endothelial CAMs,46 which in turn mediate the movement of peripheral leukocytes into the sites of inflammation. Earlier in vivo studies revealed higher myometrial expression of ICAM-1, VCAM-1 and E-selectin during parturition, mainly in the endothelium;11,47 however, the cause of such activation was not explored. We postulated that stretch-induced cytokines will upregulate expression of CAMs on uterine microvascular endothelial cells which will subsequently enhance TEM of peripheral leukocytes. Our current results confirm that media conditioned by stretched myometrial cells was able to up-regulate CAMs on human uterine microvascular ECs, likely due to a synergistic effect of multiple cytokines. Further, our results are consistent with previously described mechanism of immune cell extravasation in other systems such as inflammatory heart disease.45 We assume that with the enhanced CAM expression on ECs, the likelihood of circulating leukocytes to interact with the activated endothelium greatly increases. We should acknowledge however that in addition to stretch, soluble factors (e.g., complement48) and sex steroids49 that have been shown to modulate CAM expression also likely contribute to the increase in immune cell invasion near term.

Moreover, during extravasation, peripheral leukocytes must be activated to express integrin molecules (i.e., CD11b, also known as ITGAM) on their surfaces for interaction with the endothelium.29 CD11b is an important mediator of leukocyte recruitment; its surface expression has been found to be upregulated on peripheral monocytes and neutrophils of term pregnant women.4,5 ICAM-1 is a major ligand for leukocyte integrins LFA-1 and Mac-1 (CD11b/CD18 or amb2), whereas E-selectin binds to P-selectin glycoprotein ligand-1, hyaluronic acid receptor CD44 and Mac-1 on leukocytes.50 The a4b1 integrin, the ligand for VCAM-1, is found on the surface of monocytes, lymphocytes, eosinophils, basophils47 and neutrophils during chronic inflammation.51 In addition, CD44 (which is constitutively expressed by immune cells) was shown to be involved in the recruitment of monocytes and neutrophils to the site of inflammation.52,53 CD44–hyaluronic acid interaction could be induced by inflammatory stimuli such as cytokines.54 CD44 protein was recently found to be increased on peripheral leukocytes from term pregnant women.55 Given the crucial role in leukocyte–endothelial interaction, we hypothesized that CD11b and CD44 can mediate the recruitment of leukocytes during labour, the process stimulated by the stretch-induced release of multiple cytokines by the myometrium. In support of this hypothesis, we found that CD11b and CD44 expression on both leukocyte subtypes was increased after incubating with media conditioned by myometrial SMCs and that neutrophil adhesion to the underlying endothelial monolayer was significantly enhanced after ECs were stimulated with S-CM. Overall, this may indicate that myometrial-secreted cytokines could potentially be responsible for the induction of endothelial CAMs, integrins and CD44 on peripheral leukocytes, and could subsequently enhance leukocyte adhesion to activated endothelium during TL (the effect that we also demonstrated through adhesion assays), which promotes their infiltration into uterine tissues. It was shown by others that the adhesion of leukocytes initiate intracellular signalling pathways that increase permeability and facilitate the TEM of leukocytes.11,56,57 Leukocytes can undertake two routes during TEM: transcellular (through cytoplasm) or paracellular (between adjacent ECs) pathways.57–59 In this study, we demonstrated that myometrial cytokines (irrespective of stretch) augment vascular permeability, which allow selective yet specific passage of blood cells and macromolecules. Altogether, our current results showing the enhanced human leukocyte trafficking towards stretch-induced myometrial cytokines/chemokines in vitro support their direct involvement in the infiltration of immune cells into the uterus during labour that was reported in vivo.13,56 We suggest a model by which leukocyte recruitment to the uterus, as part of the initiation of labour, is achieved (Figure 5). According to this model, (i) mechanical stretch imposed by the growing fetus at late gestation contributes to the release of multiple myometrial cytokines, which activate (ii) peripheral maternal leukocytes and (iii) endothelium (iv) to promote their adhesion to ECs which (v) facilitate the entry of peripheral leukocytes into Cellular & Molecular Immunology

Mechanical stretch of myometrium recruits leukocytes YH Lee et al

10

2

1

CAM

3

Leukocyte Activation

Cytokine secretion

Adhesion

4

Transendothelial Migration

Inflammatory Signal Amplification

Uterine Stretch

Monocyte

5

Neutrophil

Labour

Macrophage

CAMs

Cytokines

Myometrial cell

Endothelial cell

Figure 5 Putative model of human labour. Our data suggested that (1) mechanical stretch of the myometrium by the growing fetus induces the secretion of multiple cytokines and chemokines near term. We also demonstrated that stretch-induced cytokines (2) activate peripheral leukocytes and (3) enhance the expression of endothelial CAMs. This resulted in (4) enhanced adhesion of immune cells to myometrial vascular endothelial cells, promoting (5) leukocyte transendothelial migration into the uterine muscle. Within the muscle, immune cells differentiate, producing more cytokines/chemokines, and contribute to the amplification of inflammatory signal that ultimately leads to the onset of labour and facilitates the process of postpartum involution. CAM, cell adhesion molecule.

the uterine tissues. As a result, the leukocyte infiltrate secretes more pro-inflammatory cytokines that may contribute to the amplification of inflammatory signal that ultimately leads to the onset of labour and facilitates the process of postpartum involution. ACKNOWLEDGEMENTS The authors confirm that there are no conflicts of interest. This study was supported by a grant from the CIHR (#37775). We gratefully thank Mrs Anna Dorogin for her assistance with THP-1 chemotaxis assay.

1 Golightly E, Jabbour HN, Norman JE. Endocrine immune interactions in human parturition. Mol Cell Endocrinol 2011; 335: 52–59. 2 Unal ER, Cierny JT, Roedner C, Newman R, Goetzl L. Maternal inflammation in spontaneous term labor. Am J Obstetr Gynecol 2011; 204: 223.e1–223.e5. 3 Mittal P, Romero R, Tarca AL, Gonzalez J, Draghici S, Xu Y et al. Characterization of the myometrial transcriptome and biological pathways of spontaneous human labor at term. J Perinat Med 2010; 38: 617–643. 4 Yuan M, Jordan F, McInnes IB, Harnett MM, Norman JE. Leukocytes are primed in peripheral blood for activation during term and preterm labour. Mol Hum Reprod 2009; 15: 713–724. 5 Luppi P, Irwin TE, Simhan H, Deloia JA. CD11b expression on circulating leukocytes increases in preparation for parturition. Am J Reprod Immunol 2004; 52: 323–329. Cellular & Molecular Immunology

6 Gomez-Lopez N, Tanaka S, Zaeem Z, Metz GA, Olson DM. Maternal circulating leukocytes display early chemotactic responsiveness during late gestation. BMC Pregnancy Childbirth 2013; 13(Suppl 1): S8. 7 Shynlova O, Lee YH, Srikhajon K, Lye SJ. Physiologic uterine inflammation and labor onset: integration of endocrine and mechanical signals. Reprod Sci 2013; 20: 154–167. 8 Loudon JAZ. Mechanical stretch of human uterine smooth muscle cells increases IL-8 mRNA expression and peptide synthesis. Mol Hum Reprod 2004; 10: 895–899. 9 Young A, Thomson AJ, Ledingham M, Jordan F, Greer IA, Norman JE. Immunolocalization of proinflammatory cytokines in myometrium, cervix, and fetal membranes during human parturition at term. Biol Reprod 2002; 66: 445–449. 10 Maehara K, Kanayama N, Maradny EE, Uezato T, Fujita M, Terao T. Mechanical stretching induces interleukin-8 gene expression in fetal membranes: a possible role for the initiation of human parturition. Eur J Obstet Gynecol Reprod Biol 1996; 70: 191–196. 11 Ledingham MA, Thomson AJ, Jordan F, Young A, Crawford M, Norman JE. Cell adhesion molecule expression in the cervix and myometrium during pregnancy and parturition. Obstetr Gynecol 2001; 97: 235– 242. 12 Keski-Nisula LT, Aalto ML, Kirkinen PP, Kosma VM, Heinonen ST. Myometrial inflammation in human delivery and its association with labor and infection. Am J Clin Pathol 2003; 120: 217–224. 13 Hamilton S, Oomomian Y, Stephen G, Shynlova O, Tower CL, Garrod A et al. Macrophages infiltrate the human and rat decidua during term and preterm labor: evidence that decidual inflammation precedes labor. Biol Reprod 2012; 86: 39. 14 Osmers RG, Blaser J, Kuhn W, Tschesche H. Interleukin-8 synthesis and the onset of labor. Obstetr Gynecol 1995; 86: 223–229.

Mechanical stretch of myometrium recruits leukocytes YH Lee et al

11 15 Esplin MS, Peltier MR, Hamblin S, Smith S, Fausett MB, Dildy GA et al. Monocyte chemotactic protein-1 expression is increased in human gestational tissues during term and preterm labor. Placenta 2005; 26: 661–671. 16 Kemp B, Winkler M, Maas A, Maul H, Ruck P, Reineke T et al. Cytokine concentrations in the amniotic fluid during parturition at term: correlation to lower uterine segment values and to labor. Acta Obstet Gynecol Scand 2002; 81: 938–942. 17 Christiaens I, Zaragoza DB, Guilbert L, Robertson SA, Mitchell BF, Olson DM. Inflammatory processes in preterm and term parturition. J Reprod Immunol 2008; 79: 50–57. 18 Tattersall M, Engineer N, Khanjani S, Sooranna SR, Roberts VH, Grigsby PL et al. Pro-labour myometrial gene expression: are preterm labour and term labour the same? Reproduction 2008; 135: 569–579. 19 Srikhajon K, Shynlova O, Domerudee P, Chanrachakul B, Lye SJ. A new role for monocytes in modulating myometrial inflammation during human labor. Biol Reprod 2013. Published online before print May 14, 2014, doi: 10.1095/biolreprod.113.114975 20 Gruden G, Setti G, Hayward A, Sugden D, Duggan S, Burt D et al. Mechanical stretch induces monocyte chemoattractant activity via an NF-kappaB-dependent monocyte chemoattractant protein-1mediated pathway in human mesangial cells: inhibition by rosiglitazone. J Am Soc Nephrol 2005; 16: 688–696. 21 Yamamoto H, Teramoto H, Uetani K, Igawa K, Shimizu E. Cyclic stretch upregulates interleukin-8 and transforming growth factorbeta1 production through a protein kinase C-dependent pathway in alveolar epithelial cells. Respirology 2002; 7: 103–109. 22 Kendal-Wright CE. Stretching, mechanotransduction, and proinflammatory cytokines in the fetal membranes. Reprod Sci 2007; 14(8 Suppl): 35–41. 23 Shynlova O, Nedd-Roderique T, Li Y, Dorogin A, Nguyen T, Lye SJ. Infiltration of myeloid cells into decidua is a critical early event in the labour cascade and post-partum uterine remodelling. J Cell Mol Med 2013; 17: 311–324. 24 Shynlova O, Tsui P, Dorogin A, Lye SJ. Monocyte chemoattractant protein-1 (CCL-2) integrates mechanical and endocrine signals that mediate term and preterm labor. J Immunol 2008; 181: 1470– 1479. 25 Mosser DM, Edwards JP. Exploring the full spectrum of macrophage activation. Nat Rev Immunol 2008; 8: 958–969. 26 Condon J, Yin S, Mayhew B, Word RA, Wright WE, Shay JW et al. Telomerase immortalization of human myometrial cells. Biol Reprod 2002; 67: 506–514. 27 Gra¨bner R, Till U, Heller R. Flow cytometric determination of Eselectin, vascular cell adhesion molecule-1, and intercellular cell adhesion molecule-1 in formaldehyde-fixed endothelial cell monolayers. Cytometry 2000; 40: 238–244. 28 Fiuza C, Salcedo M, Clemente G, Tellado JM. Granulocyte colonystimulating factor improves deficient in vitro neutrophil transendothelial migration in patients with advanced liver disease. Clin Diagn Lab Immunol 2002; 9: 433–439. 29 Sadik CD, Kim ND, Luster AD. Neutrophils cascading their way to inflammation. Trends Immunol 2011; 32: 452–460. 30 Grainger DJ, Reckless J. Broad-spectrum chemokine inhibitors (BSCIs) and their anti-inflammatory effects in vivo. Biochem Pharmacol 2003; 65: 1027–1034. 31 Hua R, Pease JE, Sooranna SR, Viney JM, Nelson SM, Myatt L et al. Stretch and inflammatory cytokines drive myometrial chemokine expression via NF-kB activation. Endocrinology 2012; 153: 481–491. 32 Osman I, Young A, Ledingham MA, Thomson AJ, Jordan F, Greer IA et al. Leukocyte density and pro-inflammatory cytokine expression in human fetal membranes, decidua, cervix and myometrium before and during labour at term. Mol Hum Reprod 2003; 9: 41–45. 33 Keelan JA, Blumenstein M, Helliwell RJ, Sato TA, Marvin KW, Mitchell MD. Cytokines, prostaglandins and parturition—a review. Placenta 2003; 24(Suppl A): S33–S46. 34 Friebe-Hoffmann U, Chiao JP, Rauk PN. Effect of IL-1beta and IL-6 on oxytocin secretion in human uterine smooth muscle cells. Am J Reprod Immunol 2001; 46: 226–231.

35 Rauk PN, Friebe-Hoffmann U, Winebrenner LD, Chiao JP. Interleukin-6 up-regulates the oxytocin receptor in cultured uterine smooth muscle cells. Am J Reprod Immunol 2001; 45: 148–153. 36 Romano M, Sironi M, Toniatti C, Polentarutti N, Fruscella P, Ghezzi P et al. Role of IL-6 and its soluble receptor in induction of chemokines and leukocyte recruitment. Immunity 1997; 6: 315–325. 37 Bollopragada S, Youssef R, Jordan F, Greer I, Norman J, Nelson S. Term labor is associated with a core inflammatory response in human fetal membranes, myometrium, and cervix. YMOB 2009; 200: 104.e1–104.e11. 38 Elliott CL, Slater DM, Dennes W, Poston L, Bennett PR. Interleukin 8 expression in human myometrium: changes in relation to labor onset and with gestational age. Am J Reprod Immunol 2000; 43: 272–277. 39 Greenhalgh DG, Sprugel KH, Murray MJ, Ross R. PDGF and FGF stimulate wound healing in the genetically diabetic mouse. Am J Pathol 1990; 136: 1235–1246. 40 Wengner AM, Pitchford SC, Furze RC, Rankin SM. The coordinated action of G-CSF and ELR1CXC chemokines in neutrophil mobilization during acute inflammation. Blood 2007; 111: 42–49. 41 Luppi P, Haluszczak C, Trucco M, Deloia JA. Normal pregnancy is associated with peripheral leukocyte activation. Am J Reprod Immunol 2002; 47: 72–81. 42 Baumann R, Casaulta C, Simon D, Conus S, Yousefi S, Simon HU. Macrophage migration inhibitory factor delays apoptosis in neutrophils by inhibiting the mitochondria-dependent death pathway. FASEB J 2003; 17: 2221–2230. 43 Ferrara N, Gerber HP, LeCouter J. The biology of VEGF and its receptors. Nat Med 2003; 9: 669–676. 44 Lee TH, Avraham H, Lee SH, Avraham S. Vascular endothelial growth factor modulates neutrophil transendothelial migration via upregulation of interleukin-8 in human brain microvascular endothelial cells. J Biol Chem 2002; 277: 10445–10451. 45 Golias C, Tsoutsi E, Matziridis A, Makridis P, Batistatou A, Charalabopoulos K. Review. Leukocyte and endothelial cell adhesion molecules in inflammation focusing on inflammatory heart disease. In Vivo 2007; 21: 757–769. 46 Bevilacqua MP, Pober JS, Mendrick DL, Cotran RS, Gimbrone MA. Identification of an inducible endothelial–leukocyte adhesion molecule. Proc Natl Acad Sci USA 1987; 84: 9238–9242. 47 Winkler M, Kemp B, Fischer DC, Ruck P, Rath W. Expression of adhesion molecules in the lower uterine segment during term and preterm parturition. Microsc Res Tech 2003; 60: 430–444. 48 Kilgore KS, Shen JP, Miller BF, Ward PA, Warren JS. Enhancement by the complement membrane attack complex of tumor necrosis factoralpha-induced endothelial cell expression of E-selectin and ICAM-1. J Immunol 1995; 155: 1434–1441. 49 Cutini PH, Campelo AE, Agriello E, Sandoval MJ, Rauschemberger MB, Massheimer VL. The role of sex steroids on cellular events involved in vascular disease. J Steroid Biochem Mol Biol 2012; 132: 322–330. 50 Zarbock A, Ley K. Neutrophil adhesion and activation under flow. Microcirculation 2009; 16: 31–42. 51 Canalli AA, Proenca RF, Franco-Penteado CF, Traina F, Sakamoto TM, Saad ST et al. Participation of Mac-1, LFA-1 and VLA-4 integrins in the in vitro adhesion of sickle cell disease neutrophils to endothelial layers, and reversal of adhesion by simvastatin. Haematologica 2011; 96: 526–533. 52 Cuff CA, Kothapalli D, Azonobi I, Chun S, Zhang Y, Belkin R et al. The adhesion receptor CD44 promotes atherosclerosis by mediating inflammatory cell recruitment and vascular cell activation. J Clin Invest 2001; 108: 1031–1040. 53 Khan AI, Kerfoot SM, Heit B, Liu L, Andonegui G, Ruffell B et al. Role of CD44 and hyaluronan in neutrophil recruitment. J Immunol 2004; 173: 7594–7601. 54 Pure´ E, Cuff CA. A crucial role for CD44 in inflammation. Trends Mol Med 2001; 7: 213–221. 55 Sabra S, Shynlova O, Lye SJ. Immunophenotyping of maternal peripheral blood detects activated leukocyte subpopulations Cellular & Molecular Immunology

Mechanical stretch of myometrium recruits leukocytes YH Lee et al

12 associated with preterm labor. Reprod Sci 2011; 18(3 Suppl): 269A. 56 Aghajanian A, Wittchen ES, Allingham MJ, Garrett TA, Burridge K. Endothelial cell junctions and the regulation of vascular permeability and leukocyte transmigration. J Thromb Haemost 2008; 6: 1453–1460. 57 Dejana E. Endothelial cell–cell junctions: happy together. Nat Rev Mol Cell Biol 2004; 5: 261–270.

Cellular & Molecular Immunology

58 Haidari M, Zhang W, Chen Z, Ganjehei L, Warier N, Vanderslice P et al. Myosin light chain phosphorylation facilitates monocyte transendothelial migration by dissociating endothelial adherens junctions. Cardiovasc Res 2011; 92: 456–465. 59 Yuan SY, Shen Q, Rigor RR, Wu MH. Neutrophil transmigration, focal adhesion kinase and endothelial barrier function. Microvasc Res 2012; 83: 82–88.