Microarray-based gene expression profiling reveals the mediators and pathways involved in the anti-arthritic activity of Celastrus-derived Celastrol

Abstract

Rheumatoid arthritis (RA) is an autoimmune disease characterized by chronic inflammation of the joints. The prolonged use of non-steroidal anti-inflammatory drugs and other newer drugs is associated with severe adverse reactions. Therefore, there is a need for newer anti-arthritic agents. Celastrol, a bioactive component of the Chinese herb Celastrus, possesses anti-arthritic activity as tested in the adjuvant arthritis (AA) model of rheumatoid arthritis (RA). However, the mechanism of action of Celastrol has not been fully defined. We reasoned that microarray analysis of the lymphoid cells of Celastrol-treated arthritic animals might provide vital clues in this regard. We isolated total RNA of the draining lymph node cells (LNCs) of Celastrol-treated (Tc) and vehicle-treated (Tp) arthritic Lewis rats that were restimulated in vitro with the disease-related antigen, mycobacterial heat-shock protein 65 (Bhsp65), and tested it using microarray gene chips. Also tested was RNA from LNCs of control arthritic rats just before any treatment (T₀). Seventy six genes involved in various biological functions were differentially regulated by Bhsp65 in LNCs of Tp group, and 19 genes among them were shared by the Tc group. Furthermore, a group of 14 genes was unique to Tc. When Tc and Tp were compared, many of the Bhsp65-induced genes were related to the immune cells, cellular proliferation and inflammatory responses. Our results revealed 10 differentially expressed genes and 14 pathways that constituted the "Celastrol Signature". Our results would help identify novel targets for RA therapy.

Figure 1. The expression level of Bhps65-induced DEG identified in LNC before and after treatment with Celastrol and their categorization according to GO analysis

A total of 32 function-defined DEG involved in immune response (A), cellular activation and proliferation (B), oxygen metabolism (C), lipid metabolism or other metabolic processes (D), and other functions (E) are shown.

Figure 2. Comparison of Bhsp65-induced gene expression profiles of PBS-treated and Celastrol-treated arthritic rats

(A) heatmap of Bhsp65-induced DEGs in LNC of PBS-treated (left) and Celastrol-treated (right) rats with AA. Color bars represent the expression level of DEGs. Venn Diagrams show the numbers of DEGs individually in PBS- or Celastrol-treated group as well as the overlapping DEG in these two groups. ↑, upregulation of gene expression compared to the corresponding baseline (gene expression of LNC cultured in medium); ↓,downregulation of gene expression compared to the baseline. (B) The main functional categories of 64 (PBS-treated)/28 (Celastrol-treated) functionally well-defined DEG. Dark color shows downregulated genes, whereas light color represents the upregulated genes in each category. (The group of “non-defined genes” shown in Table 1 is not included in Figure 2B)

Figure 3. Correlation of gene expression obtained from microarray and from qPCR

X axis, gene expression determined by RatRef-12Expression BeadChip (Illumina) hybridization; Y axis, gene expression determined by quantitative PCR. Data is represented as -fold over medium after normalization. (r2=0.7069, p <0.0001)

Figure 4. Comparative Bio-function analysis of Tc and Tp groups

Data sets were analyzed by the IPA software (Ingenuity® Systems) is expressed as a p-value (< 0.05), which was calculated using the right-tailed Fisher’s Exact Test. Figure shows the top 10 (most significant) bio-functional groups. Threshold is set at p = 0.05.

Figure 5. Comparative Canonical Pathway analysis

Data sets were analyzed and compared by ‘IPA’. The significance is expressed as a p-value (< 0.05), which was calculated using the right-tailed Fisher’s Exact Test. Figure shows the top 10 (most significant) pathways in each group and their counterparts in the other group. P = 0.05 served as threshold.