PBMCs and monocyte isolation

Blood was collected from healthy controls and patients either on the day of catheterization, or within 3 months of the date of echocardiography, in CPT tubes designed for one-step cell separation (Becton Dickinson). The sample was then immediately mixed and centrifuged at 1,800g at ambient temperature for 30 minutes. Positive selection was performed to isolate monocytes. After PBMC isolation, monocytes were magnetically labeled with CD14, and CD14+ and CD14− (flow-through) cell populations were isolated using a MACS MicroBeads Column (Miltenyi Biotec).

RNA isolation and microarray analysis hybridization

Total RNA from PBMCs, enriched monocytes retained on the column, and nonmonocyte cells (flow-through from the column) was extracted using RNeasy Mini kits (Qiagen). RNA was isolated using RNeasy minicolumns according to the protocol of the manufacturer (Qiagen). RNA samples were stored at −80°C.

Microarray data clustering

All microarray data from this study have been deposited in the GEO (Gene Expression Omnibus) database of the National Center for Biotechnology Information (http://www.ncbi.nlm.nih.gov/geo/; accession no. GSE19617). Microarray data were clustered using Cluster 3.0 for Mac OSX (C Clustering Library 1.43), using hierarchical clustering. After selecting genes showing detectable expression in 80% of the samples, genes and then arrays were centered and normalized before clustering by gene.

Quantitative reverse transcription–polymerase chain reaction (RT-PCR)

Complementary DNA was synthesized from total RNA using reverse transcriptase (Gibco BRL) and random primers (Applied Biosystems) as described previously (12). RT-PCR was performed using an ABI StepOne Plus sequence detection system (Applied Biosystems) and TaqMan assay, according to an Applied Biosystems amplification protocol (User Bulletin 2: http://www3.appliedbiosystems.com/cms/groups/mcb_support/documents/generaldocuments/cms_040980.pdf). The comparative threshold method was used for data analysis, and 18S ribosomal RNA (rRNA) was used as control (the relative expression of the target RNA was calculated in relation to its values). Primers for 18S rRNA and messenger RNA (mRNA) of the following target genes were obtained from Applied Biosystems: IL13Ra1 (Hs00609817_m1), CCR1 (Hs00174298_m1), JAK2 (Hs00234567_m1), MX1 (Hs00895598_m1), Siglec1 (Hs00988063_m1), and MRC1 (Hs00267207_m1).

Stimulation of PBMCs from healthy controls and SSc patients

PBMCs were plated in 24- or 48-well plates at 37°C, at a concentration of 1–2 × 10⁶ cells per well, in complete medium (RPMI 1640 supplemented with 10% fetal bovine serum, 20 mM L-glutamine, 100 IU/ml penicillin, and 100 gm/ml streptomycin). Stimulation consisted of coculture with one of the following reagents: lipoteichoic acid (1 μg/ml) as the Toll-like receptor 2 (TLR-2) agonist, poly(I-C) (25 μg/ml) as the TLR-3 agonist, ultrapure LPS derived from Escherichia coli O111:B4 (1 μg/ml) as the TLR-4 agonist, imiquimod R837 (5 μg/ml) as the TLR-7 agonist, and CpGA–oligodeoxy-nucleotide 2216 (1 μg/ml) as the TLR-9 agonist (all from InvivoGen). PBMCs were then incubated with or without (negative control) addition of recombinant human IL-13 (20 ng/ml; R&D Systems) for 18 hours and then either lysed in the Buffer RLT (Qiagen) for RNA preparation and analysis as described above or stained for flow cytometric analyses (see below).

Flow cytometry

PBMCs were labeled with fluorescein isothiocyanate–conjugated mouse anti-human CD14 monoclonal antibody (BD PharMingen), allophycocyanin–Cy7–conjugated mouse anti-human CD206 (anti-MRC1) (BioLegend), and goat IgG anti-human IL-13 receptor antagonist type I (IL-13Ra1) antibody (R&D Systems) used with phycoerythrin-conjugated donkey anti-goat IgG (Jackson ImmunoResearch). Immunofluorescence was measured with a FACScan flow cytometer (BD Biosciences), and the data were analyzed with FlowJo software (Tree Star). The data are expressed as the percent of cells positive for each population.

IL-4 and IL-13 assay

Plasma was collected in EDTA blood collection tubes, aliquoted, and stored at −80°C. Plasma levels of human IL-13 and human IL-4 were analyzed by enzyme-linked immunosorbent assay (Thermo Scientific), including calibration curves using supplied assay diluents for human plasma with a range of 0–1,000 pg/ml.

IFN-regulated and PAH biomarker genes expressed by PBMCs in lcSSc patients cluster into distinct groups

Using our previously described microarray database (3), genes were clustered using average linkage, hierarchical, and supervised clustering, grouping healthy subjects, lcSSc patients without PAH, and lcSSc patients with PAH. This clustering showed increased expression of a cluster of IFN-regulated genes in selected lcSSc patients (Figure 1A). This cluster of genes is similar to that previously reported by our group in patients with dcSSc (4) and was found both in patients with PAH and in those without PAH. The IFN-regulated genes found in this cluster include many well-known IFN-regulated genes: IFN-induced with helicase C domain 1 I (IFIH1), myxovirus (influenza virus) resistance (MX1), IFN-induced protein with tetratrico-peptide repeat 1 (IFIT1 and IFIT2, IFN alpha-inducible proteins 2 and 3 (G1P2 and G1P3), IFI44, IFI44L, IFITM3, IFITM2, GBP1, IFN-inducible RNA-dependent protein kinase (PRKR), IFI27, IFN regulatory factor 7 (IRF7), and IFI35.

Open in a separate window

Figure 1

Biomarker and interferon (IFN)–regulated gene expression in limited cutaneous systemic sclerosis (lcSSc). A, Heat map showing the expression of genes clustered using average linkage, hierarchical, and supervised clustering. Individual subjects are listed at the top. Top right, Cluster of IFN-regulated genes; middle right, cluster of biomarker genes; bottom right, JAK2 cluster. Values above the mean expression level of each gene are indicated in red; those below the mean expression level are indicated in green. B, Relative expression of all genes in each cluster in A. Data are shown as box plots. Each box represents the 25th to 75th percentiles. Lines outside the boxes represent the minimum and maximum values. Lines inside the boxes represent the mean. lSSc-NoPAH = lcSSc without pulmonary arterial hypertension; lSSc-PAH = lcSSc with PAH.

Furthermore, we observed a distinct cluster of genes, including genes that we recently reported as biomarkers that distinguish lcSSc patients from healthy controls and lcSSc patients with PAH from those without PAH (Figure 1A). Most of the biomarker genes clustered together, including interleukin-13 receptor, alpha 1 (IL13RA1), interferon gamma receptor 1 (IFNGR1), chemokine (C-C motif) receptor 1 (CCR1), allograft inflammatory factor 1 (AIF1), tissue inhibitor of metalloproteinases 2 (TIMP2), and intercellular adhesion molecule 1 (ICAM1). Notably, JAK2, one of the genes previously identified as a biomarker gene for lcSSc with PAH, clustered separately, although it showed a similar pattern of increased expression in lcSSc with PAH. This cluster was notable for also including the JAK2 signaling partner STAT3 and multiple TLR genes (TLR1, TLR4, TLR6, and TLR7) (for a complete list of genes in each of these clusters, see Supplemental Table II, available on the Arthritis & Rheumatism Web site at http://onlinelibrary.wiley.com/journal/10.1002/(ISSN) 1529-0131).

Thus, IFN-regulated genes increased in a subset of lcSSc patients, and biomarker genes that distinguish healthy controls, lcSSc patients without PAH, and lcSSc patients with PAH clustered in groups. Although some patients, such as patients 45, 62, and 83, had up-regulated genes in both clusters, other patients had up-regulated genes in only one of these clusters, for example, patients 100 and 55 (see Figure 1A).

The average normalized gene expression in these 3 clusters was strikingly different. Whereas the biomarker cluster showed a graded increase in gene expression from healthy controls to lcSSc patients without PAH to lcSSc patients with PAH, the expression of IFN cluster genes was increased in lcSSc regardless of the presence of PAH. Notably, the JAK2 cluster of genes showed on average increased expression in lcSSc patients with PAH, but not in lcSSc patients without PAH or controls (Figure 1B).

Increased expression of IFN-regulated genes in monocytes and nonmonocyte cells from both lcSSc patients and dcSSc patients

Siglec1 is an IFN-regulated monocyte/macrophage marker that we previously showed to be highly expressed by the monocytes and tissue macrophages of dcSSc patients, and its expression in skin correlates with the MRSS (4,13). Increased expression of Siglec1 on PBMCs was observed in all SSc subtypes compared with healthy controls (P = 0.02) (Figure 2a). Expression of MX1, another gene highly induced by IFN but also up-regulated in nonmonocyte cell types (14), was also found to be higher in SSc patients than in healthy controls (P = 0.03), regardless of the SSc subtype (Figure 2c). MX1 and Siglec1 expression correlated positively (r = 0.35, P = 0.02).

Open in a separate window

Figure 2

IFN-regulated gene expression in lcSSc and diffuse cutaneous SSc (dcSSc, or dSSc). a and c, Expression of Siglec1 (a) and MX1 (c) in peripheral blood mononuclear cells (PBMCs) from healthy controls (HC), dcSSc patients, lcSSc patients without PAH, lcSSc patients with PAH, and patients with idiopathic PAH (IPAH). b and d, Expression of Siglec1 (b) and MX1 (d) in PBMC, CD14−, and CD14+ populations in lcSSc patients without PAH (n = 4), lcSSc patients with PAH (n = 5), and healthy controls (n = 3) (lcSSc patients without PAH and lcSSc patients with PAH are combined). Data are expressed as the fold change normalized to mRNA expression in a single sample from a healthy control subject. Bars show the mean ± SEM. # and ## = P < 0.05 versus all SSc patients. See Figure 1 for other definitions.

For both Siglec1 and MX1, increased expression was seen in dcSSc patients, lcSSc patients without PAH, and lcSSc patients with PAH, without any significant difference between these 3 patient groups. In contrast, MX1 and Siglec1 were not highly expressed on PBMCs from patients with idiopathic PAH but had expression levels similar to those in the healthy control group (P > 0.05 for both) (Figures 2a and c). In order to better understand the importance of monocytes in the IFN cluster, we isolated CD14+ cells from the PBMCs of 9 lcSSc patients (4 without PAH and 5 with PAH) and 3 healthy controls. Siglec1 was expressed almost exclusively by CD14+ monocytes, whereas MX1 was found on both CD14+ and CD14− cell populations (Figures 2b and d), indicating that this signature is not limited to monocytes. The expression of Siglec1 and MX1 mRNA by PBMCs did not correlate with the MRSS or pulmonary artery pressure (data not shown).

Increased expression of biomarker genes for lcSSc with PAH by PBMCs from SSc patients

Three highly overexpressed genes from the lcSSc with PAH biomarker clustering were selected for further investigation using RT-PCR on PBMCs from dcSSc patients, lcSSc patients without PAH, lcSSc patients with PAH, patients with idiopathic PAH, and healthy controls. These genes included IL13RA1, a receptor required for most actions of the Th2 cytokine IL-13 (15); CCR1, a target chemokine receptor of Th2 cytokines such as CCL6, CCL13, and CCL18 released by macrophages upon IL-13 or IL-4 stimulation and essential in mediating IL-13-induced inflammation (15,16); and JAK2, an intracellular signaling protein activated by IL-13 (17) that clustered in a group different from other biomarker genes (Figure 1A).

As we reported recently, lcSSc patients with PAH had on average higher mRNA expression of these 3 genes compared with lcSSc patients without PAH (Figures 3a, c, and e). PBMCs from lcSSc patients with PAH showed higher expression of CCR1 and JAK2 compared with PBMCs from dcSSc patients (P < 0.05 and P < 0.005, respectively) and compared with PBMCs from lcSSc patients without PAH (P < 0.05 for both). In contrast, IL13RA1 was found to be more highly expressed in all 3 subtypes than in healthy controls, but showed higher expression in dcSSc patients and lcSSc patients with PAH compared with controls (P < 0.05 for both). In contrast, PBMCs from patients with idiopathic PAH showed expression of these 3 genes similar to that in PBMCs from healthy controls (P > 0.05 for all) (Figures 3a, c, and e). Notably, although these genes fall into a cluster that is distinct from the IFN cluster, IFN-regulated Siglec1 expression correlated positively and significantly with IL13RA1 expression (r = 0.44, P < 0.005) and JAK2 expression (r = 0.35, P = 0.03). Similarly, IFN-regulated MX1 expression also correlated positively with IL13RA1 expression (r = 0.32, P = 0.02) but not with JAK2 expression (r = 0.39, P = 0.07). CCR1 expression correlated with both IL13RA1 expression (r = 0.63, P < 0.001) and JAK2 expression (r = 0.83, P < 0.001).

Open in a separate window

Figure 3

PAH biomarker gene expression in diffuse cutaneous SSc (dcSSc, or dSSc) and lcSSc. a, c, and e, Expression of IL13RA1 (a), CCR1 (c), and JAK2 (e) in peripheral blood mononuclear cells (PBMCs) from healthy controls (HC), dcSSc patients, lcSSc patients without PAH, lcSSc patients with PAH, and patients with idiopathic PAH (IPAH). b, d, and f, Expression of IL13RA1 (b), CCR1 (d), and JAK2 (f) in PBMC, CD14−, and CD14+ populations in lcSSc patients without PAH (n = 4), lcSSc patients with PAH (n = 5), and healthy controls (n = 3) (lcSSc patients without PAH and lcSSc patients with PAH are combined). Data are expressed as the fold change normalized to mRNA expression in a single sample from a healthy control. Bars show the mean ± SEM. # = P < 0.001, by analysis of variance. See Figure 1 for other definitions.

As was the case for Siglec1 (Figure 2b), IL13RA1 and CCR1 were expressed almost exclusively by CD14+ monocytes both in lcSSc patients (4 without PAH and 5 with PAH) and in healthy controls (Figures 3b and d), while JAK2 gene expression was found in both CD14+ and CD14− cell populations (Figure 3f). IL13RA1, CCR1, and JAK2 mRNA expression did not correlate with the MRSS or with PAP (data not shown).

Increased expression of MRC1, a marker of macrophage alternative activation, in lcSSc with PAH

Elevated IL13RA1 expression, mainly in the CD14+ cell population, suggested that alternative activation might contribute to monocyte activation in SSc. We therefore analyzed MRC1 expression in SSc PBMCs as a marker of alternative activation. MRC1 expression was highly and exclusively up-regulated in lcSSc patients with PAH compared with dcSSc patients, lcSSc patients without PAH, patients with idiopathic PAH, or healthy controls (overall P < 0.001) (Figure 4a). Additionally, MRC1 was the only biomarker studied whose expression correlated with PAP on catheterization of lcSSc patients (r = 0.52, P = 0.03) (Figure 4b). This positive correlation persisted if only lcSSc patients with PAH were analyzed (r = 0.61, P = 0.04). Those lcSSc patients with PAH who had higher PBMC MRC1 expression also had higher mortality during the 2-year followup period than did lcSSc patients without PAH (P = 0.02).

Open in a separate window

Figure 4

MRC1 expression in SSc patient populations and MRC1 expression by monocytes. a, Expression of MCR1 in peripheral blood mononuclear cells (PBMCs) from healthy controls (HC), patients with diffuse cutaneous SSc (dcSSc, or dSSc), lcSSc patients without PAH, lcSSc patients with PAH, and patients with idiopathic PAH (IPAH). Shaded circles represent lcSSc patients with PAH who died during the 2 years of followup in this study. b, Linear regression analysis of the relationship between the mean pulmonary artery pressure (PAP) by right-heart catheterization and the expression of MRC1 in PBMCs from lcSSc patients with PAH. c, Expression of CD14 on PBMC, CD14−, and CD14+ populations from SSc patients and controls. d, Expression of MRC1 on PBMC, CD14−, and CD14+ populations in lcSSc patients without PAH (n = 4), lcSSc patients with PAH (n = 5), and healthy controls (n = 3) (lcSSc patients without PAH and lcSSc patients with PAH are combined). e, Expression of Siglec1, MRC1, and IL13RA1 on PBMCs from healthy controls (n = 7) after 18 hours of stimulation with 20 ng/ml interleukin-13 (IL-13) or without IL-13 stimulation (control). f, Plasma IL-13 concentration in healthy controls, dcSSc patients, lcSSc patients without PAH, and lcSSc patients with PAH. Data in a and c-e are expressed as the fold change normalized to mRNA expression in a single sample from a healthy control. Bars show the mean ± SEM. # = P < 0.001, by analysis of variance. ELISA = enzyme-linked immunosorbent assay (see Figure 1 for other definitions).

MRC1 is more highly expressed in CD14+ cells and is induced by IL-13

In order to clarify the cell type responsible for increased MRC1 expression, we analyzed MRC1 mRNA expression by CD14+ monocytes from lcSSc patients and controls. CD14+ cell isolation was effective, as shown by CD14 expression in the fractionated cell populations (Figure 4c), and MRC1 expression in these samples was higher in CD14+ cells (Figure 4d), although MRC1 was also expressed at higher levels in CD14− cells from lcSSc patients than in those from healthy controls. To confirm previously reported findings, we tested the effect of IL-13 on PBMCs from 7 healthy controls and found dramatic up-regulation of MRC1 expression after 18 hours of IL-13 stimulation, with no effect on Siglec1 or IL13RA1 expression (Figure 4e).

IL-13 in lcSSc patients with PAH

IL-13 was analyzed in the plasma of lcSSc patients and healthy controls. The highest concentration of IL-13 was found in lcSSc patients with PAH (P < 0.001) (Figure 4f). Plasma IL-4 levels were undetectable in samples from lcSSc patients or healthy controls (data not shown).

IL-13 induces surface expression of MRC1 but not IL13RA1 by CD14+ monocytes

In order to further investigate whether higher mRNA levels of MRC1 are more likely related to IL-13 induction or are instead a reflection of increased IL13RA1 expression, PBMCs from an lcSSc patient with PAH were isolated and stimulated with IL-13. CD14, MRC1 (CD206), and IL13RA1 were analyzed by flow cytometry. MRC1 was highly induced in CD14+ cells after IL-13 treatment (by 4.81% before treatment and by 12.8% after treatment) (Figure 5a). As shown in Figure 5b, despite detectable baseline expression of IL13RA1 on the majority of PBMCs (51.5%), no significant induction of IL13RA1 was observed after IL-13 stimulation (53.5%), in contrast to increased PBMC expression of MRC1 after IL-13 stimulation. The high MRC1 induction and the contrasting absence of IL13RA1 expression on PBMCs after IL-13 stimulation were confirmed by RT-PCR (Figure 5c).

Open in a separate window

Figure 5

MRC1 (CD206) surface expression on peripheral blood mononuclear cells (PBMCs) from lcSSc patients with PAH. A, Double staining (CD14+CD206+) of PBMCs before (4.81%) and after (12.8%) stimulation with interleukin-13 (IL-13). B, Induction of IL13RA1 and CD206 surface expression on all PBMCs before (red lines) and after (blue lines) 3 hours of stimulation with 20 ng/ml IL-13. After IL-13 stimulation, positive cell expression of IL13RA1 changed from 51.5% (red) to 53.5% (blue), and positive cell expression of CD206 changed from 8.45% (red) to 15.3% (blue). Horizontal bars represent stained cells. C, Expression of IL13RA1 MRC1 and IL13RA1 mRNA before and after IL-13 stimulation. D, MRC1 mRNA expression on healthy PBMCs stimulated for 24 hours with IL-13 or with several Toll-like receptor (TLR) ligands. Data are expressed as the fold change normalized to mRNA expression in a single sample from a healthy control. See Figure 1 for other definitions.

ROLE OF THE STUDY SPONSOR

Actelion Pharmaceuticals was not involved in the study design, data collection, data analysis, or writing of the manuscript, and publication of the manuscript was not contingent upon their approval.

Supported by Actelion Pharmaceuticals (grant to Dr. Lafyatis) and by the NIH (National Institute of Arthritis and Musculoskeletal and Skin Diseases grant U01-AR-055063 to Dr. Lafyatis). Dr. Pendergrass’ work was supported by the NIH (National Institute of Arthritis and Musculoskeletal and Skin Diseases T32 training grant AR-007575-11). Ms Affandi’s work was supported by the Dutch Arthritis Association (grant NR-10-1-301). Dr. Whitfield’s work was supported by the Arthritis Foundation (Hulda Irene Duggan Arthritis Investigator Award). Dr. Farber’s work was supported by the Scleroderma Foundation.