The Rheumatoid Arthritis Risk Gene LBH Regulates Growth in Fibroblast-like Synoviocytes

Abstract

Objective: Fibroblast-like synoviocytes (FLS) are key players in the synovial pathology of rheumatoid arthritis (RA). Currently, there is no treatment that specifically targets these aggressive cells. By combining 3 different "omics" data sets, i.e., 1) risk genes in RA, 2) differentially expressed genes, and 3) differential DNA methylation in RA versus osteoarthritis (OA) FLS, we identified LBH (limb bud and heart development) as a candidate gene in RA. The present study was undertaken to define the role of this gene in FLS.

Methods: Synovial tissue specimens from RA and OA patients were collected at the time of joint replacement surgery. LBH expression was silenced using small interfering RNA or overexpressed using an LBH expression vector in primary FLS. Gene expression profiles were determined by microarray and assessed using Ingenuity Pathway Analysis. Effects of modified LBH expression were investigated in functional assays.

Results: LBH was expressed in the synovial lining layer in patients with RA. Transforming growth factor β1 significantly increased LBH expression in primary FLS, and platelet-derived growth factor BB decreased it. Pathway analysis of the transcriptome of LBH-deficient FLS compared to control FLS identified "cellular growth and proliferation" as the most significantly enriched pathway. In growth assays, LBH deficiency increased FLS proliferation. Conversely, LBH overexpression significantly inhibited cell growth. Cell cycle analysis demonstrated a marked increase in cells entering the cell cycle in LBH-deficient FLS compared to controls. LBH did not alter apoptosis.

Conclusion: LBH is a candidate gene for synovial pathology in RA. It is regulated by growth factors and modulates cell growth in primary FLS. Our data suggest a novel mechanism for synovial intimal hyperplasia and joint damage in RA.

Figure 1

LBH methylation in rheumatoid arthritis (RA) fibroblast-like synoviocytes. The location of the LBH RA single-nucleotide polymorphism (SNP) identified by genome-wide association study (GWAS) and the methylation levels of each of the CpGs on the bead array that are within the LBH promoter region (−2,500 bp to +500 bp from the transcription start site [TSS] [arrow]), and the transcript variant numbers of the RefSeq genes that are transcribed from that TSS are noted. Values are the mean ± SEM. ★= significantly differentially methylated CG (RA versus osteoarthritis [OA] and normal [NL]).

Figure 2

LBH expression in synovial tissue and fibroblast-like synoviocytes (FLS). A and B, Immunohistochemical staining of human synovial tissue from a rheumatoid arthritis (RA) patient, using anti-LBH antibody (A) and control normal rabbit Ig (B). Arrowhead indicates the lining layer. Bars = 50 µm. C, Expression of LBH in 3 primary RA FLS and in 1 FLS line transfected with control small interfering RNA (siRNA) and LBH siRNA, assessed by Western blotting. The membrane was probed with anti–β-actin antibody and with anti-LBH antibody as indicated. D, Expression of LBH mRNA in whole synovial tissue specimens from osteoarthritis (OA) and RA patients, assessed by quantitative polymerase chain reaction. C_t values were normalized to GAPDH expression. Each symbol represents an individual sample; bars show the mean. REU = relative expression units.

Figure 3

Effects of transforming growth factor β1 (TGFβ1) and platelet-derived growth factor (PDGF) on LBH expression in fibroblast-like synoviocytes (FLS). A and C, Expression of LBH mRNA by serum-starved rheumatoid arthritis (RA) FLS without stimulation or after stimulation for 6 hours with recombinant human TGFβ1 at various concentrations (0.1–10 ng/ml) (A) or for 12 hours with PDGF-BB at various concentrations (0.1–10 ng/ml) (C). LBH mRNA expression was measured by quantitative polymerase chain reaction. B and D, Time course of stimulation with TGFβ1 (10 ng/ml) (B) or with PDGF-BB (10 ng/ml) (D). The experiments were repeated 3 times, and C_t values were normalized to GAPDH expression. Values are the mean ± SEM of 3–8 different FLS lines. * = P < 0.05 versus no stimulation. REU = relative expression units.

Figure 4

Genes affected by LBH silencing and overexpression. A, Heatmap showing probes that were significantly differentially expressed between control (C), LBH^low (knockdown [KD]), and LBH^high (overexpression [O]) fibroblast-like synoviocytes (FLS). Analyses of differential expression included all genes for which the adjusted P value was <0.05 in any of the 3 comparisons (LBH^high versus control, LBH^high versus LBH^low, and control versus LBH^low). To make the patterns of differential expression clearer, gene expression levels are represented by a Z score, which was calculated individually for each gene. B, The top-ranked network discovered with Ingenuity Pathway Analysis (IPA) during assessment of LBH^low versus control FLS. The gene expression, cellular development, and cellular growth and proliferation functions were assigned to the network by IPA. Genes that were repressed during knockdown are shown in blue, and those that were induced are shown in red. Genes with P values of <0.01 were used in the analysis. The network contained 43 genes, of which 29 passed the cutoff.

Figure 5

Effects of modified LBH expression on cell growth in fibroblast-like synoviocytes (FLS). A, Rheumatoid arthritis (RA) FLS lines were transfected with control or LBH small interfering RNA (siRNA) (mean gene silencing 91%). The cells were replated on day 1 posttransfection and subjected to MTT assay on days 1, 3, and 7. Values are the mean ± SEM absorbance of 3 different FLS lines. LBH deficiency significantly increased cell growth (P = 0.007 versus controls). B, RA FLS lines were transfected with control or LBH expression vector (mean overexpression 13-fold). The cells were recultured on day 3 posttransfection and subjected to MTT assay on days 3, 5, and 7. Values are the mean ± SEM absorbance of 6 different FLS lines. LBH overexpression significantly decreased cell growth (P = 0.005 versus controls). C, RA FLS were transfected with control or LBH expression vector. The cells were replated on day 3 posttransfection, and caspase 3/7 activity, as a measure of apoptosis, was determined on day 5. Fluorescence values were normalized to cell number, measured by MTT. Values are the mean±SEM of 3 different RA FLS lines.

Figure 6

Cell cycle progression in fibroblast-like synoviocytes (FLS) after LBH gene silencing. Rheumatoid arthritis (RA) FLS lines were transfected with control or LBH small interfering RNA (siRNA) and cultured until subconfluent. The cells were harvested and fixed on day 4 posttransfection and, after DNA staining, were subjected to flow cytometry. A, Representative histograms of DNA content in RA FLS lines transfected with control siRNA or LBH expression vector. The number of cells in each phase of the cell cycle was estimated using Watson Pragmatic modeling. B, Mean ± SEM percentage of cells in the S + G₂/M phases in 3 different RA FLS lines. * = P = 0.001.