Novel application of multi-stimuli network inference to synovial fibroblasts of rheumatoid arthritis patients

Abstract

Background: Network inference of gene expression data is an important challenge in systems biology. Novel algorithms may provide more detailed gene regulatory networks (GRN) for complex, chronic inflammatory diseases such as rheumatoid arthritis (RA), in which activated synovial fibroblasts (SFBs) play a major role. Since the detailed mechanisms underlying this activation are still unclear, simultaneous investigation of multi-stimuli activation of SFBs offers the possibility to elucidate the regulatory effects of multiple mediators and to gain new insights into disease pathogenesis.

Methods: A GRN was therefore inferred from RA-SFBs treated with 4 different stimuli (IL-1 β, TNF- α, TGF- β, and PDGF-D). Data from time series microarray experiments (0, 1, 2, 4, 12 h; Affymetrix HG-U133 Plus 2.0) were batch-corrected applying 'ComBat', analyzed for differentially expressed genes over time with 'Limma', and used for the inference of a robust GRN with NetGenerator V2.0, a heuristic ordinary differential equation-based method with soft integration of prior knowledge.

Results: Using all genes differentially expressed over time in RA-SFBs for any stimulus, and selecting the genes belonging to the most significant gene ontology (GO) term, i.e., 'cartilage development', a dynamic, robust, moderately complex multi-stimuli GRN was generated with 24 genes and 57 edges in total, 31 of which were gene-to-gene edges. Prior literature-based knowledge derived from Pathway Studio or manual searches was reflected in the final network by 25/57 confirmed edges (44%). The model contained known network motifs crucial for dynamic cellular behavior, e.g., cross-talk among pathways, positive feed-back loops, and positive feed-forward motifs (including suppression of the transcriptional repressor OSR2 by all 4 stimuli.

Conclusion: A multi-stimuli GRN highly concordant with literature data was successfully generated by network inference from the gene expression of stimulated RA-SFBs. The GRN showed high reliability, since 10 predicted edges were independently validated by literature findings post network inference. The selected GO term 'cartilage development' contained a number of differentiation markers, growth factors, and transcription factors with potential relevance for RA. Finally, the model provided new insight into the response of RA-SFBs to multiple stimuli implicated in the pathogenesis of RA, in particular to the 'novel' potent growth factor PDGF-D.

Figure 1

Impact of the parameter variations. Influence of the NetGenerator parameter ‘allowedError’ on average MSE (+), number of network edges (o), and number of integrated prior knowledge edges (△). The optimized model, selected on the basis of low average MSE and high number of integrated prior knowledge edges (indicated by a dotted line), showed an average MSE of 2.91, 17 integrated prior knowledge edges, 84 network edges in total, and an ‘allowedError’of 0.045.

Figure 2

Time courses of measured and simulated gene expression data. Each panel displays the results for one of the 24 differentially expressed genes (DEGs) selected from GO term ‘cartilage development’, comparing measured and simulated expression values (both in a scaled form) over time (h). The measured, interpolated data are indicated by dashed lines, the simulated expression data by solid lines, with each color representing one of the 4 stimuli (IL-1 β = turquoise; TNF- α green; TGF- β = red, and PDGF-D = purple).

Figure 3

Consensus network of the 24 differentially expressed genes (DEGs). The model contains nodes representing the 4 stimuli and the 24 selected DEGs. The heuristic optimization leads to an optimal fit of the model to the measured data and is preferably based on inferred edges supported by prior knowledge (represented in green). Edges ‘externally’ validated by additional knowledge are emphasized by green double-line edges. However, the model also contains edges only predicted from the expression data (represented in black) and one edge conflicting with prior knowledge (represented in red).