High ICAM-1 gene expression in pulmonary fibroblasts of COPD patients: a reflection of an enhanced immunological function

Subjects

To investigate mRNA expression, lung tissue was obtained from 20 individuals. The clinical characteristics of the groups are presented in table 1⇓. Informed consent was given by all patients. Classification of severity of COPD was based on the 2003 GOLD criteria 4, 20. Three groups were studied: 1) GOLD stage II/III (n = 7); this group included individuals with moderate (GOLD stage II, n = 5) and severe COPD (GOLD stage III, n = 2). 2) GOLD stage IV (n = 7); this group contained individuals with very severe COPD. 3) Control (n = 6); this group included individuals with histologically normal lungs. Patients had no α₁-antitrypsin deficiency. Emphysema was defined as histologically abnormal enlargement of the airspaces distal to the terminal bronchioles, accompanied by destruction of alveolar walls and without obvious fibrosis 1, 2, 4, and was assessed by routine histological examination of lung tissue by an experienced pulmonary pathologist.

GOLD stage II/III tissue was obtained from noninvolved lung tissue from patients undergoing resective surgery for pulmonary carcinoma. Tissue was always taken as far away as possible from the tumour, or from a noninvolved lobe. Histopathologically, the emphysematous lesions were present, yet lesions were of limited but varying severity. Moderate emphysema was noted histopathologically as isolated or unattached segments of viable alveolar septal tissue or isolated cross-sections of pulmonary vessels.

Tissue from the very severe COPD group (GOLD stage IV) was obtained from patients with COPD undergoing surgery for lung transplantation or lung volume reduction. The resected tissue showed both macroscopically and microscopically severe emphysematous lesions, often accompanied by bullae.

The control group consisted of donor lung tissue that had not been used for transplantation because of technical or logistical (unilateral transplantation) reasons. Technical reasons were related to doubt as to technical transplantability in relation to available recipients or doubts concerning the condition of the lung. In case of the latter, lung tissue was included only after additional investigations (including histopathology), and when it showed no signs of COPD 1, or of any other significant pathology.

Isolation and culture of lung fibroblasts

Pulmonary fibroblast cultures were established from parenchymal lung tissue by means of an explant technique as described previously 11. Absence of mycoplasma contamination in the fibroblast cultures was confirmed with a mycoplasma detection kit (Roche Diagnostics, Almere, the Netherlands). Isolated cells were characterised as fibroblasts by morphological appearance and expression pattern of specific proteins 11. All cells exhibited a characteristic staining pattern for vimentin, fibronectin, and the fibroblast marker prolyl-4-hydroxylase and lacked immunoreactivity for keratin. It was found that <5% of cells were positive for desmin and α-smooth muscle cell actin.

Supernatants

Epithelial supernatant was obtained from a bronchial epithelial cell line, 16HBE 21 (kind gift from D.C. Gruenert, University of Vermont, California Pacific Medical Center Research Institute, San Francisco, CA, USA). The cell line 16HBE was grown to confluence in complete medium (Ham's F12 medium; BioWhittaker Cambrex, Verviers, Belgium) supplemented with 10% foetal calf serum (FCS; PAA laboratories, Linz, Austria), 2 mM L-glutamine, 100 µg·L⁻¹ streptomycin and 100 U·mL⁻¹ penicillin (BioWhittaker Cambrex). After confluence, culture medium was replaced with medium containing 0.5% FCS for 24 h. This supernatant was harvested and stored at −80°C.

To obtain culture supernatant of CD8+ T-cells, peripheral blood was collected from seven patients with COPD. Informed consent was given by all patients. Heparinised blood was obtained and layered on Ficoll-paque (Amersham Biosciences, Uppsala, Sweden). The interface containing peripheral blood mononuclear cells (PBMCs) was collected. PBMCs were resuspended at 1×10⁷ cells·mL⁻¹ in Hank's balanced salt solution/2% FCS for purification of CD8 cells. CD8 cells were isolated by positive selection using anti-CD8 magnetic beads and Detach-a-bead (Dynal, Wirral, UK). The purity of the isolated CD8 T-cells was determined by fluorescence-activated cell sorter analysis on a coulter Epics ELITE (Beckman Coulter, Hialeah, FL, USA) with CyQ-labelled anti-CD8, phycoerythrin labelled anti-CD4 and fluorescein isothiocyanate-labelled anti-CD3 antibodies (BD Bioscience, Oxford, UK). The CD8 cells were 87% pure containing 7% CD4 T-cells. Isolated CD8 cells were resuspended in culture medium containing 0.5% FCS and incubated for 24 h. Isolated CD8 cells were stimulated with anti-CD3 and anti-CD28 antibodies by replacing the culture medium with 0.5% FCS medium containing 5% anti-CD3 (clone WT-32) and anti-CD28 antibodies (clone 20–4996; CLB Sanquin, Amsterdam, the Netherlands) for 24 h. Supernatant was then harvested, centrifuged and stored at −80°C.

Fibroblast culture for cDNA arrays

Fibroblasts from confluent cultures were seeded in culture plates (Corning BV, Amsterdam, the Netherlands) in complete medium and grown to confluence (passage 5). Culture medium was replaced with medium containing 0.5% FCS for 24 h. Medium was replaced with new 0.5% FCS medium, either with or without supernatant of the 16HBE cell line (1 in 25 dilution), supernatant of CD8 cells stimulated with anti-CD3 and anti-CD28 (1 in 10 dilution) or a combination of supernatant of 16HBE cells and CD8 cells (1 in 25 and 1 in 10 final dilution concentration, respectively). Optimal stimulation concentration and duration was determined in pilot experiments (data not shown). After 6 h cells were harvested for RNA isolation.

Preparation of RNA and hybridisation of cDNA arrays

Fibroblast total RNA was isolated using the RNeasy mini kit (Qiagen, Hilden, Germany). RNA quantity was determined by optical density measurements at 260 nm. A total of 3 µg RNA per patient was pooled per group. Total RNA was treated with DNase I and run over a column to remove genomic DNA (Qiagen).

The Atlas Cell Interaction expression array (BD Biosciences Clontech, Palo Alto, CA, USA) was used to identify differences in the expression pattern of 265 known genes in the fibroblasts from the three patient groups treated with supernatants of 16HBE cells, CD8 cells, or their combination. The Atlas expression arrays are based on a filter array technique (macro-array) that uses a nylon filter spotted with cDNAs of the genes of interest that can be hybridised with cDNA of the samples to be analysed. A total of 2–5 µg total RNA was used for cDNA synthesis using a commercial kit (BD Biosciences). The cDNA was labelled by incorporation of [α-³²P] deoxycytidine triphosphate (10 µCi·µL⁻¹). The probes were hybridised when the counts·min⁻¹ were 2–8×10⁶, with the ExpressHyb hybridisation solutions applied to the cDNA arrays following the user manual (BD Biosciences). The membranes were exposed to a phosphorimaging screen at room temperature for 3 days. Quantification of bound cDNA was performed using Phosphorimaging system (Molecular Dynamics, Sunnyvale, CA, USA).

Analysis of cDNA array data

The current authors compared gene expression patterns of fibroblasts from the three groups after incubation with supernatant of 16HBE cells, CD8 cells, or the combination and at basal culture conditions, using AtlasImage 2.0 software (BD Biosciences Clontech). In order to subtract the common variance and find the differentially expressed genes, a new method was designed based on the data reduction tool “principal component analysis” 22–24.

Briefly, a matrix was prepared with all filter data placed in columns. To each value, 1 was added to avoid values of zero. The natural logarithm was taken to improve normality according to the following formula:

All columns were standardised to mean = 0 and sd = 1. Subsequently, the principal components were calculated and the first principal component was deducted from the data matrix. Of the remaining data, the inverse of the natural logarithm was taken. Ratios were determined and a gene expression ratio of at least a two-fold increase compared with basal culture conditions was considered relevant.

Real time PCR

To obtain a more uniform patient group the only patients who had longstanding bronchitis (COPD stage IV) were excluded for PCR and subsequently for the soluble intercellular adhesion molecule (sICAM) ELISA and immunohistochemistry. To validate the results of the array data, quantitative real time PCR analysis was performed for ICAM-1 mRNA expression on individual patient fibroblast samples used in the array study. A total of 3 ng of mRNA was transcribed into cDNA by reverse transcriptase II (Invitrogen, Breda, the Netherlands). Real time PCR was performed on an ABI7900HT sequencer with Assay on Demands from Applied Biosystems (Foster City, CA, USA), according to manufacturer's instructions. Data were analysed by the ΔΔCT method 25. In brief, CT values of the genes of interest were corrected for CT values from a household gene (ribosomal protein S9), resulting in a ΔCT value. This ΔCT was normalised to the average ΔCT value of the basal control group. The log value was taken and ratios between stimulated and basal condition were calculated. Medians were taken from the individual patients per category.

sICAM protein expression

Of all fibroblasts samples, supernatants were harvested to determine sICAM levels produced by fibroblasts. sICAM levels were determined by a commercial sICAM ELISA kit (CLB/Sanquin, Amsterdam, the Netherlands). In brief, wells were coated with a capturing sICAM antibody and subsequently incubated with the supernatant samples. After incubation, wells were washed and incubated with a peroxidase labelled anti-sICAM antibody. Staining was visualised using 3,3',5,5'-tetramethylbenzidine. Absorption was read at 450 nm on a standard photospectrometer.

Immunohistochemistry

Immunohistochemistry was performed on 3 µm formalin fixed, paraffin embedded lung tissue using a standard two-step immunoperoxidase protocol. Sections were incubated with a primary monocolonal antibody recognising ICAM-1 (kind gift from A.W. Boyd, Royal Melbourne Hospital, Melbourne, Australia) 26. Envision (Dakopatts, Glostrup, Denmark) in combination with 3,3'-diaminobenzidine (Sigma, St. Louis, MO, USA) was used to visualise antibody binding. Slides were semiquantitatively scored for the number of ICAM-1+ cells in the submucosa. (1: few ICAM-1+ cells, 2: moderate number of ICAM-1+ cells, 3: large number of ICAM-1+ cells).

Statistical analysis

Differences in subject characteristics, real-time data and immunohistochemistry scores between the three study groups were analysed using the Kruskall–Wallis test followed by the Mann–Whitney U-test. Differences in smoking habits and sex between the three study groups were analysed using the Fisher's exact test. The level of significance used was 0.05. All reported p-values are two-sided.

Patient groups

Pack-yrs smoking histories were similar in subjects with moderate/severe and very severe COPD (table 1⇑). Patients with very severe COPD were significantly younger than patients with moderate COPD. Material from the current study's stage II group mainly consisted of males, material of the stage IV group mainly consists of females. From the present authors'archives, it was not possible to obtain material that was equally divided in males and females per COPD stage. The control group did not significantly differ in age or sex distribution from the COPD groups. Lung material from disapproved donor lungs was used as a control group. It was not possible to obtain all clinical characteristics of the donors, as could be obtained form the current study's patient group. However, this material was analysed by an experienced pathologist and classified as normal tissue.

Array results in general

A total of 51 out of 265 genes tested were differentially expressed on the cell interaction arrays. Genes were categorised into three groups: genes involved in adhesion and migration, genes involved in ECM production or modulation and genes involved in cellular regulation. Overall, cell-interaction genes were especially differentially expressed in the control and stage IV group. The stage II/III group showed only minor changes in gene expression profiles.

Gene expression with epithelium supernatant stimulation

After epithelial supernatant stimulation, 22 out of 265 genes were differentially expressed when compared with fibroblasts under basal culture conditions (table 2⇓); in particular genes involved in adhesion and regulation. Epithelial supernatant influenced predominantly gene expression in control fibroblasts, but had hardly any effect on COPD fibroblasts. One gene in particular was differentially expressed: ICAM-1 expression was downregulated (−4.4 times) in the control group and, in contrast, upregulated in both COPD groups (4.5- and 5.7-fold in stage II/III and stage IV, respectively).

Gene expression after CD8 supernatant stimulation

A total of 29 genes were differentially expressed upon CD8 supernatant stimulation (table 2⇑), with 23 genes being differentially expressed in the GOLD stage IV COPD patient group and only nine and three genes expressed in GOLD stage II/III and control groups, respectively. Most regulated genes reflected adhesion processes and ECM substances. Notably ICAM-1 expression was highly upregulated in both COPD groups (10.2- and 9.5-fold in GOLD stage II/III and GOLD stage IV, respectively) whereas the control group showed no regulation of ICAM-1 mRNA expression.

Gene expression after combined CD8 and epithelial supernatant

After stimulation with a combination of CD8 and epithelial supernatants, 24 genes were differentially regulated (table 2⇑). Fibroblasts of controls treated with the combined supernatants primarily showed alteration in the expression of genes involved in cell regulatory processes. Fibroblasts of GOLD stage II/III COPD patients showed minimal alteration under these conditions whereas in the GOLD stage IV COPD group, ICAM-1 showed a surprisingly high increase in mRNA expression (25.5-fold).

Real time PCR

Arrays were used to find indications for which genes would be most fit for further analysis of genes of interest by real time PCR. ICAM-1 expression was the most prominent change found in the arrays to explore and confirm by RT-PCR. The current authors found the highest ICAM-1 expression in GOLD stage IV fibroblasts under the influence of the epithelial and CD8 supernatant combination (fig. 1⇓). Epithelial supernatant alone also significantly elevated the ICAM-1 gene expression in GOLD stage IV compared with GOLD stage II and controls. The CD8 supernatants also caused the highest level of ICAM-1 mRNA expression in the GOLD stage IV group, but this was not significant, due to large patient variation in combination with the relatively small number of subjects.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 1—

Real time intercellular adhesion molecule-1 mRNA expression of fibroblasts stimulated with supernatants of a) epithelial and CD8 cells, b) epithelial cells or c) CD8 cells. Ratio compared with basal culture conditions. GOLD: Global Initiative for Chronic Obstructive Lung Disease. ––: median. ^#: p =  −0.028; ^¶: p =  −0.028. One GOLD stage IV sample was lost due to technical reasons.

sICAM protein expression

To verify the mRNA expression, sICAM levels in the supernatants of all fibroblast samples were also determined, using an sICAM ELISA. Very small concentrations of sICAM were detected under basal conditions. Stimulation with the combination of CD8 and epithelial supernatant showed a trend towards an increase in sICAM levels in GOLD stage IV fibroblasts (p = 0.055, fig. 2⇓). Epithelial and CD8 supernatant stimulation alone did not affect the protein levels of sICAM.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 2—

Soluble intercellular adhesion molecule-1 protein expression of fibroblasts stimulated with supernatants of a) epithelial and CD8 cells, b) epithelial cells or c) CD8 cells. Ratio compared with basal culture conditions. GOLD: Global Initiative for Chronic Obstructive Lung Disease. ^#: p = 0.055.

Immunohistochemistry

To further verify the highly elevated mRNA expression of GOLD stage IV fibroblasts, ICAM-1 protein expression in tissue sections of the same control and COPD patients from whom fibroblasts were derived, as used in the array study, were also analysed. In the submucosa, where a large number of fibroblasts are located and is likely to be influenced by local epithelial and CD8-derived factors, a significantly larger number of ICAM-1+ cells were found in GOLD stage IV COPD patients when compared with controls and GOLD stage II/III COPD patients (fig. 3⇓). Most positive cells represented elongated fibroblast-like cells (fig. 4⇓). This supports the present authors' observation of the high increase in ICAM-1 mRNA expression in GOLD stage IV COPD fibroblasts.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 3—

Semi-quantitative score of number of intercellular adhesion molecule (ICAM-1)+ cells in the submucosa. Relative frequency of ICAM-1+ cells is significantly increased in Global Initiative for Chronic Obstructive Lung Disease (GOLD) stage IV chronic obstructive pulmonary disease.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 4—

Intercellular adhesion molecule (ICAM)-1 staining (brown) in lung tissue of a) control, b) Global Initiative for Chronic Obstructive Lung Disease (GOLD) stage II COPD and c) GOLD stage IV COPD subjects. In the control, as well as GOLD stage II COPD, very small numbers of ICAM-1+ cells were observed in the submucosa. Arrows indicate elongated ICAM-1 negative fibroblasts in the submucosa. In GOLD stage IV COPD, numerous ICAM-1+ cells were observed. Arrows indicate the ICAM-1+ elongated fibroblasts. Cells were counterstained with haematoxylin.