Fibroblasts profiling in scarring trachoma identifies IL-6 as a functional component of a fibroblast-macrophage pro-fibrotic and pro-inflammatory feedback loop

Trachoma is a conjunctiva scarring disease, which is the leading infectious cause of blindness worldwide. Yet, the molecular mechanisms underlying progressive fibrosis in trachoma are unknown. To investigate the contribution of local resident fibroblasts to disease progression, we isolated conjunctival fibroblasts from patients with scarring trachoma and matching control individuals, and compared their gene expression profiles and functional properties in vitro. We show that scarring trachoma fibroblasts substantially differ from control counterparts, displaying pro-fibrotic and pro-inflammatory features matched by an altered gene expression profile. This pro-inflammatory signature was exemplified by increased IL-6 expression and secretion, and a stronger response to macrophage-mediated stimulation of contraction. We further demonstrate that scarring trachoma fibroblasts can promote Akt phosphorylation in macrophages in an IL-6 –dependent manner. Overall this work has uncovered a distinctive molecular fingerprint for scarring trachoma fibroblasts, and identified IL-6- as a potential contributor to the chronic conjunctival fibrosis, mediating reciprocal pro-fibrotic/pro-inflammatory interactions between macrophages and fibroblasts.


Biopsy collection and expansion
Biopsies from cases were collected from adults (age >18 years old) with clearly established Trachomatous Trichiasis with associated scarring of the upper tarsal conjunctiva (Grade 2 or 3 on the detailed WHO FPC Trachoma grading scale), undergoing trichiasis surgery. Patients that had been subjected to previous surgery or had conjunctival scarring from a cause other than trachoma were excluded from the study. Control biopsies were collected from adults without Trachomatous Trichiasis or any clinically apparent conjunctival disease or scarring, which were due to undergo ophthalmic surgery for a condition unrelated to disease of the conjunctiva (eg cataract) and had no previous record of eyelid surgery.
The eyelid was anaesthetized with an injection of 2% lignocaine and the eye cleaned with 5% povidone iodine. A biopsy sample was taken using a 3mm trephine from the tarsal conjunctiva, 2mm from the lid margin, at the junction of the medial and lateral of the everted lid. None of these cases had evidence of current C. trachomatous infection as detected by PCR (Amplicor PCR, Roche). The biopsies were either wrapped in sterile gauze and moistened with normal saline or placed in Optimem medium (Life Technologies) and transported to the laboratory at +8°C. Biopsies derived from trachoma and control patients were used for fibroblast expansion, with similar age range and gender for both sets (Table S2). The explants where placed in 32 mm tissue culture plates after being dissected and digested at 37 o C for 10-15min using 0.05% Collagenase in PBS. Cells were then cultured in Dulbecco's modified Eagle's medium (DMEM) (PAA laboratories, Austria or Life Technologies) supplemented with 10% (v/v) heat-inactivated foetal bovine serum (FBS, Sigma-Aldrich, UK), 100 IU/ml penicillin, 100 μg/ml streptomycin (hereafter referred to as complete medium) until 40% confluence and then trypsinised and passaged in a T25 tissue culture flask (Corning, NY, USA). Contaminating epithelial and goblet cells were eliminated from the cultures after the first or second passage. Cultures were assessed for typical fibroblast morphology by phase contrast microscopy before every experiment. Cells were then maintained in complete medium and used between passages 2 and 8 for all the experiments. Serum-free medium was used where indicated with or without the addition of cytokines.

Collagen Contraction assay
Control and diseased fibroblasts were embedded in three-dimensional, collagen type I matrix (First Link Ltd., Birmingham, UK) at a final collagen concentration of 1.4 mg/ml after pH was rapidly adjusted to 7 using 1M NaOH. Cells were added to the collagen mixture at a final concentration 8 x10 4 cells/ml or 16 x 10 4 /ml for FBSstimulated or serum free contraction assays, respectively. For co-culture experiments, macrophages were added to the gel mix at a final ratio of 1 (fibroblast): 4 (macrophages) (8 x10 4 cells/ml : 3.2x10 5 cells/ml) and cultured in 10% FBS or SF DMEM medium. The collagen lattices were cast on Mattek® dishes (MatTek Corporation, Ashland, MA) and after 20 min incubation at 37°C gels were detached and medium was added. Reduction in lattice area at days 1, 3, and 7 due to contraction was digitally photographed, and the gel areas calculated using image analysis software (ImageJ, rsbweb.nih.gov/ij/).

Collagen gel imaging
Immunofluorescence was performed as described before (1). Briefly, collagen gels were fixed using 3.7% Formaldehyde (Sigma-Aldrich, UK) followed by a 30-minute incubation in 0.5% Triton-X100 (Sigma) and 0.1 M glycine. The gels were then stained with Rhodamine-phalloidin (Invitrogen, Life Technologies, UK) in Tris Buffer Saline (TBS) additioned with 1% BSA and 1% FBS in TBS at various time points during contraction. The gels were then imaged using a Biorad Radiance confocal microscope (Zeiss Axiovert S100/Biorad Radiance 2000; Zeiss, Cambridge, UK) to visualize the cells (red, Green HeNe laser 540⁄565 nm) and the matrix (confocal reflection microscopy) using a long working distance objective (ZEISS LD plan-Neofluoar 63x/0.75).

Elastic modulus calculation
The Elastic Modulus (Young's modulus) of the tissue contracts was calculated from the formula: !" = !×! , where !" is the matrix force and ! is the cross-sectional area of the gels (2) with ! = !"##$% ℎ!"#ℎ × !"#$% !"#$%ℎ. Tissue height was evaluated manually using the focus drive function on a software driven (OpenLab, Improvision) Zeiss Axiovert 100M microscope (Zeiss plan-neofluar 10x/0.30). The length of the probe used for the indentations was 5 mm. For each tissue construct, tissue height was averaged from 5 measurements, and 4 gels per experiment and per condition were used to calculate the elastic modulus.

Flow cytometry
Flow cytometry was used to assess integrin expression of 3 Control and 3 trachoma cell lines. Cells were trypsinised, washed with PBS and incubated with FITC conjugated antibodies for 1h on ice. The conjugated antibodies used are listed in Table   S1. Cells were acquired on a BD FACSCalibur using CellQuest (BD Cytometry Systems) at the FACS facility in the Institute of Ophthalmology, UCL. Flow cytometry analyses were performed using FlowJo Software (Tree star, OR USA).

Macrophage differentiation
U937 monocytes were cultured in RPMI1640 (Gibgo, Life technologies, UK) supplemented with 10% FBS, 100 IU/ml penicillin, and 100 μg/ml streptomycin (complete RPMI). For differentiation into macrophages, the cells were incubated with 100 nM PMA for 3 days in complete RPMI, and further rested for 2-3 days in regular complete RPMI.

Metabolic activity
Cell metabolic activity was measured at day 4 of fibroblasts growing in collagen gels.
Cells/gels were incubated for 3 hour in 37 0 C with 10% Alamar blue in 10% FBS DMEM medium. After 3 hours supernatants were examined for fluorescence absorbance, according to manufacture guidelines.

MMP activity
MMP activity released in the medium during gel contraction was measured at day 5 using an MMP activity kit (MMP Activity Assay Kit, Abcam, UK), following manufacturers instructions.

Pro-collagen synthesis
An enzyme immunoassay (TaKara BIO INC.) was used to determine the Pro-collagen Type I C-peptide concentration in conditioned medium of contracting gels. Supernatants from Day 3 and 7 of contraction were collected and kept in -80 o C until used. The supernatants from 8 control and 8 cases were examined for Pro-collagen Type I C-peptide concentration, following manufacturer's protocol for higher sensitivity Samples. The % change in pro-collagen I synthesis was calculated for each cell line.

Real time qPCR
Reverse transcription was carried out from the extracted mRNA using the QuantiTect Reverse Transcription Kit (Qiagen, UK) according to manufacturer's instructions. qRT-PCR reactions were performed on a real-time PCR system (HT7900 Fast Real-Time PCR; Applied Biosystems). Data were analysed using the comparative ΔCT method (3).