Transcriptomic analysis in LPS-stimulated microglia cocultured with rBM-

The effects of rBM-MSCs on LPS-stimulated microglia were also determined via reversed conditions of the aforementioned coculture system in vitro. To that end, microglia were seeded on the bottom chamber as subject groups and subjected to 3 different conditions: control, LPS stimulation, and LPS stimulation with rBM-MSC coculture (Figure 12(a)). There were no significant changes in cell density of microglia (data not shown). The transcriptome of subject groups was analysed using microarrays and IPA. Comparison between LPS stimulation and LPS stimulation with rBM-MSC coculture showed changes in expression levels of genes related to cancer, organismal injury and abnormalities, and cell death (Figure 13). Gene ontology analysis of the transcriptome revealed altered expression levels of genes related to inflammatory response related canonical pathways such as triggering receptor expressed on myeloid cell 1 (TREM1) signalling, neuroinflammation, and rheumatoid arthritis (Table 4 and Figure 12(b)). Expression of genes related to TREM1 signalling and neuroinflammation was especially suppressed, as predicted (negative z-score). Focused gene expression analysis of the inflammatory response showed significantly altered levels in 64 genes between groups (Figure 12(c)).

Although the differences were more pronounced after LPS stimulation, remarkable changes were also observed between the presence and absence of rBM-MSCs.

- 31 -

Figure 12. Inflammation-related gene expression variation in LPS-stimulated microglia cocultured with rat bone marrow-derived mesenchymal stem cells (rBM-MSCs). (a) In vitro reverse coculture experimental design. Three different conditions were used (group 1–3): group 1, control (microglia only); group 2, LPS-stimulated microglia; and group 3, LPS-stimulation with rBM-MSC coculture. (b) Canonical pathway analysis was constructed algorithmically using Ingenuity Pathway Analysis based on microarray data. Bars indicate canonical pathways containing genes with significantly altered expression. Bar graph colours from blue (inhibition) to orange (activation) represent gene activity of the corresponding pathway according to z-score. (c) Heat map of genes related to inflammation with altered expression levels. Gene expression values are coloured from green (downregulated) to red (upregulated).

- 32 -

Figure 13. Top 20 list of function or diseases constructed algorithmically by Ingenuity Pathway Analysis in LPS-stimulated microglia cocultured with rat bone marrow-derived mesenchymal stem cells (rBM-MSCs) compared to LPS-stimulated microglia.

- 33 -

Table 5. Top 20 canonical pathways constructed algorithmically by Ingenuity Pathway Analysis in LPS-stimulated microglia cocultured with rat bone marrow-derived mesenchymal stem cells (rBM-MSCs) compared to LPS-stimulated microglia

Canonical Pathways -log(p-value) Number of genes

TREM1 Signaling 11.7 31

Neuroinflammation Signaling Pathway 10.4 73

Role of Macrophages, Fibroblasts and Endothelial Cells in

Rheumatoid Arthritis 9.65 70

Dendritic Cell Maturation 8.54 49

iNOS Signaling 7.17 19

IL-10 Signaling 7.04 24

Type I Diabetes Mellitus Signaling 6.93 31

Th1 Pathway 6.87 35

Role of PKR in Interferon Induction and Antiviral Response 6.83 18

Th1 and Th2 Activation Pathway 6.82 43

PI3K Signaling in B Lymphocytes 6.45 34

Altered T Cell and B Cell Signaling in Rheumatoid Arthritis 6.31 26

Colorectal Cancer Metastasis Signaling 5.97 51

JAK/Stat Signaling 5.88 24

Role of Pattern Recognition Receptors in Recognition of

Bacteria and Viruses 5.81 33

Crosstalk between Dendritic Cells and Natural Killer Cells 5.75 25

HMGB1 Signaling 5.64 32

Production of Nitric Oxide and Reactive Oxygen Species in

Macrophages 5.51 42

Death Receptor Signaling 5.46 25

Toll-like Receptor Signaling 5.46 22

- 34 -

3.5. Functional prediction of transcriptomic networks and reduced inflammatory response in LPS-stimulated microglia cocultured with rBM-MSCs

To examine differences in transcriptomes between LPS-stimulated microglia with and without rBM-MSCs, gene expression networks related to inflammatory were generated using IPA in the corresponding groups (Figure 14 and Table 5). A total of 65 genes highly related to the inflammatory response were identified, and the genes were directly linked. Based on the differential expression (upregulation or downregulation), prediction analysis of networks showed that the inflammatory response is predicted to be inhibited by rBM-MSCs (Figure 15(a) and Table 6). Three genes with altered expression levels were highly related to inflammation, which was confirmed using qPCR (Figure 15(b)). LPS-induced upregulated levels of tumor necrosis factor (Tnf), C-C motif chemokine ligand 2 (Ccl2), and toll-like receptor 2 (Tlr2) genes were significantly decreased after coculture with rBM-MSCs. These predicted functional results were confirmed in experiments using cell cultures, where the number of activated cells induced by LPS stimulation showing swelled and round morphology was significantly decreased by rBM-MSCs (Figure 15(c)). Moreover, the levels of protein markers for microglial activation, CD40 and CD74, were also lower in the presence of rBM-MSC than after LPS stimulation without rBM-MSC, indicating that direct phenotypical activation of microglia was reduced by rBM-MSCs (Figure 16).

- 35 -

Figure 14. Transcriptomic network analysis in two different culture conditions of LPS-stimulated microglia. Gene network related to inflammatory response was constructed algorithmically using Ingenuity Pathway Analysis. Transcriptome network of (a) LPS-stimulated microglia and (b) LPS-stimulated microglia cocultured with rBM-MSCs compared to control (microglia only), respectively were shown. Red and green areas indicate up- and down-regulated genes, respectively.

Differentially expressed genes were obtained from microarray data (>1.2 fold-change).

- 36 -

Figure 15. Reduced inflammatory response in LPS-stimulated microglia cocultured with rat bone marrow-derived mesenchymal stem cells (rBM-MSCs).

(a) Gene network related to inflammatory response was constructed, and cellular function was predicted algorithmically using Ingenuity Pathway Analysis. Red and green areas indicate up- and downregulated genes, respectively. Differentially expressed genes were obtained from microarray data (>1.2 fold-change). (b) Quantitative real-time PCR analysis of gene expression-related inflammation in LPS-stimulated microglia cocultured with rBM-MSCs compared to control (microglia only). (c) Activated microglia were counted in light microscopy images and quantified as the percentage of activated microglia/total cell number. Cells at the edge of the images were not counted. Scale bar: 20 μm. ∗p < 0.05 and ∗∗p < 0.01 versus control (microglia only).

- 37 -

Table 6. 64 genes related to inflammatory response in LPS-stimulated microglia cocultured with rat bone marrow-derived mesenchymal stem cells (rBM-MSCs) compared to LPS-stimulated microglia

No. Symbol Illumina Expr Fold Change Location

1 AKT1 ILMN_1353102 -1.35 Cytoplasm

2 BIRC2 ILMN_1650704 1.47 Cytoplasm

3 BIRC3 ILMN_1365233 1.67 Cytoplasm

4 CASP1 ILMN_1361826 -1.27 Cytoplasm

5 CHUK ILMN_2039288 1.24 Cytoplasm

6 CYBB ILMN_1352424 -1.53 Cytoplasm

7 HMOX1 ILMN_1650285 -1.61 Cytoplasm

8 IRAK3 ILMN_1361037 -1.31 Cytoplasm

9 JAK1 ILMN_1354897 -1.38 Cytoplasm

10 JAK3 ILMN_1355371 1.22 Cytoplasm

11 MAPK14 ILMN_1352809 1.35 Cytoplasm

12 MAPK3 ILMN_1366612 -1.31 Cytoplasm

13 NCF2 ILMN_1365484 -1.71 Cytoplasm

14 NLRP3 ILMN_1362786 -1.66 Cytoplasm

15 NOS2 ILMN_1372107 1.38 Cytoplasm

16 PIK3CG ILMN_1351492 -1.59 Cytoplasm

17 PIK3R5 ILMN_1350193 -1.32 Cytoplasm

18 PLA2G4A ILMN_1360060 1.21 Cytoplasm

19 PLCG2 ILMN_1367941 -1.28 Cytoplasm

20 PPP3CA ILMN_1372414 1.29 Cytoplasm

21 PSEN2 ILMN_1368494 1.53 Cytoplasm

22 PTGS2 ILMN_1349422 3.05 Cytoplasm

23 PTPN11 ILMN_1363069 -1.26 Cytoplasm

24 SOD2 ILMN_1367263 -1.25 Cytoplasm

25 SYK ILMN_1350279 -1.53 Cytoplasm

26 TBK1 ILMN_1367327 -1.42 Cytoplasm

27 CCL2 ILMN_1368224 -2.44 Extracellular Space

28 CNTF ILMN_1349254 1.27 Extracellular Space

29 CXCL10 ILMN_1364335 -1.84 Extracellular Space

30 IL10 ILMN_1369848 -1.63 Extracellular Space

31 IL12B ILMN_1354970 -1.54 Extracellular Space

32 IL18 ILMN_1352725 -1.34 Extracellular Space

- 38 -

No. Symbol Illumina Expr Fold Change Location

33 IL6 ILMN_1355154 1.61 Extracellular Space

34 MFGE8 ILMN_1363839 1.57 Extracellular Space

35 NGF ILMN_1362230 1.92 Extracellular Space

36 TNF ILMN_1357947 -2.55 Extracellular Space

37 WNT1 ILMN_1358775 1.21 Extracellular Space

38 CASP8 ILMN_1371431 -1.24 Nucleus

39 CTNNB1 ILMN_1352752 -1.34 Nucleus

40 FOS ILMN_1368356 5.65 Nucleus

41 JUN ILMN_1365705 1.74 Nucleus

42 NFAT5 ILMN_1373056 1.57 Nucleus

43 NFE2L2 ILMN_1372879 -1.34 Nucleus

44 NFKB2 ILMN_1374155 1.46 Nucleus

45 STAT1 ILMN_1372159 -1.54 Nucleus

46 CD40 ILMN_1355111 -1.38 Plasma Membrane

47 CD86 ILMN_1374614 1.21 Plasma Membrane

48 HLA-DRA ILMN_1376935 -1.28 Plasma Membrane

49 ICAM1 ILMN_1354506 -1.22 Plasma Membrane

50 IFNGR1 ILMN_1366851 2.55 Plasma Membrane

51 IFNGR2 ILMN_1356831 -1.26 Plasma Membrane

52 IL6R ILMN_1349175 1.41 Plasma Membrane

53 IRAK2 ILMN_1373683 -1.65 Plasma Membrane

54 MYD88 ILMN_1359024 -1.23 Plasma Membrane

55 PPP3CB ILMN_1351926 -1.23 Plasma Membrane

56 PSENEN ILMN_1374774 1.20 Plasma Membrane

57 SLC6A12 ILMN_1353136 1.90 Plasma Membrane

58 TICAM2 ILMN_1357791 -1.21 Plasma Membrane

59 TLR2 ILMN_1353896 -2.57 Plasma Membrane

60 TLR3 ILMN_1355463 -1.30 Plasma Membrane

61 TLR4 ILMN_1364385 -1.24 Plasma Membrane

62 TLR6 ILMN_1356202 1.53 Plasma Membrane

63 TLR7 ILMN_1371182 -1.56 Plasma Membrane

64 TYROBP ILMN_1372404 1.26 Plasma Membrane

Expr Fold Change: Experimental fold change

- 39 -

(a) (b)

Figure 16. Western blot analysis in LPS-stimulated microglia cocultured with rat bone marrow-derived mesenchymal stem cells (rBM-MSCs). (a) Total Protein was extracted from microglia in three different conditions. Samples were resolved on a 15% gel and western blotting was performed using anti-CD40 and CD74 antibodies. β-actin served as a loading control. (b) The graph shows quantification of western blot bands. Data represent the mean of three independent experiments. (mean ± SD).

- 40 -

Figure 17. Schematic representation of bidirectional interaction between rat bone marrow-derived mesenchymal stem cells (rBM-MSCs) and activated microglia. Activated microglia increased movement of rBM-MSCs in rBM-MSCs cocultured with LPS-stimulated microglia in an in vitro coculture system. rBM-MSCs decreased the inflammation of activated microglia in a reversal coculture system.

- 41 -