Increased CXCL10 expression in MS MSCs and monocytes is unaffected by AHSCT

Abstract

Objective: To confirm CXCL10 over production in bone marrow mesenchymal stem cells (MSCs) and circulating monocytes isolated from multiple sclerosis patients (MS) and identify predate cell molecular signature; to extend this analysis after autologous hematopoietic stem cell transplantation (AHSCT) to test if therapy has modifying effects on MSCs and circulating monocytes.

Methods: MSCs and monocytes were isolated from 19 MS patients who undergone AHSCT before and seven of them at least 3 years after transplant. CXCL10 production was detected after LPS/IFN-γ stimulation. TLR4 signaling pathways were investigated by means of transcription factors phosphorylation/activation level. RT-PCR of activated transcription factors was performed to quantify their expression. All experiments were conducted in parallel with 24 matched healthy donors (HD).

Results: CXCL10 expression was significantly increased in both peripheral circulating monocytes and BM MSCs compared to HD. We showed that CXCL10 production is determined by an altered signaling pathway downstream TLR4, with the involvement of STAT-1, NF-κB, p38, JNK, and CREB. All upregulated transcription factors are more phosphorylated in MS patient sample. These features are not modified after AHSCT.

Interpretation: We demonstrated that in MS two different cell lineages are characterized by significantly increased production of CXCL10, due to altered signaling pathways of innate immune reaction mediated by TLR4, probably associated with disease phenotype. This characteristic is not modified by AHSCT, suggesting that when T and B lymphocytes are reset, other possible components of MS pathology, such as CXCL10 over production, do not determine therapy outcome.

Figure 1

MSCs/monocytes CXCL10 production before AHSCT. (A) MSCs from 19 multiple sclerosis (MS) patients (gray circles) and 24 healthy donors (HD, black circles) were analyzed for CXCL10 production upon LPS stimulation at three in vitro passages, P2, P4, and P6. MSCs were plated in 24-well plates at the density of 1 × 10⁵ cells/mL, incubated for 24 h with LPS 1 μg/mL. CXCL10 was measured on cell supernatants by ELISA. Each point in the graph corresponds to an independent MSC sample; horizontal bars represent CXCL10 mean value (pg/mL). Here, we show that MS MSCs produce higher level of CXCL10 with respect to controls at each tested passage. CXCL10 mean production is significantly higher in MS MSCs than HD MSCs at each passage (**P < 0.002 at P2, *P < 0.05 at P4, ***P < 0.001 at P6, nonparametric two-tailed t-test). (B) CXCL10 production was tested by real-time PCR on eight MS MSCs (gray Whiskers box plot) and eight HD (black Whiskers box plot) in basal condition (−) and upon stimulation with LPS for 4 h (+LPS). mRNA expression level is reported as ratio to β-actin. CXCL10 mean expression in MS MSCs is significantly higher than HD samples under stimulated condition (*P < 0,05, nonparametric two-tailed t-test). (C) All MS and HD MSCs included in the study were tested after LPS stimulation, for the production of the cytokines IL-1α-1β-2-4-6-10-12p40-12p70-17-23, TNFα, OPN, IFNγ, GM-CSF, MMP9, and CXCL13 by Milliplex, and for IL-8, IFNβ production by ELISA. Here, we show that MSCs produce GM-CSF, IL-8, IL-6, and OPN, without any difference between MS and HD samples (mean values, expressed as pg/mL, ±SEM are reported). (D) CXCL10 production was investigated in six MS and six HD samples of peripheral CD14+ monocytes under basal condition (−) and upon LPS stimulation (+LPS) by Milliplex. Monocytes were plated in 24-well plates at the density of 1 × 10⁶ cells/mL and then incubated for 24 h with or without LPS 1 μg/mL. Here, we show that MS monocytes produce higher level of CXCL10 than controls before and after LPS stimulation (P < 0,03, Mann–Whitney test). (E) In the same MSCs and monocytes samples analyzed as above, we evaluated CXCL10 production after IFNγ stimulation (10 ng/mL for 24 h) by ELISA. All MSCs and monocytes produce CXCL10 in response to IFNγ, without any difference between MS and HD. (F) BM plasma of seven MS patients and six HD was tested ex vivo for cytokines. The presence of IL-1α,β; IL-2-4-6-IL10-12p40-12p70-17-23, TNFα, OPN, IFNγ, GM-CSF, MMP9, and CXCL13 was evaluated by Milliplex; IL-8 and IFNβ were tested by ELISA. Among the tested cytokines, we found that BM plasma contains CXCL10, CXCL13, MMP-9, OPN, and IFNβ; MS samples contain significantly higher amounts of CXCL10 (**P = 0.01, Mann–Whitney test) and OPN (*P = 0.04, Mann–Whitney test) than HD ones. HD, healthy donors; MS, multiple sclerosis; MSCs, mesenchymal stem cells.

Figure 2

CXCL10 production after AHSCT. (A) Follow-up and clinical characteristics of seven of 19 MS patients after AHSCT. Left panel: at long-term follow-up, with a median of 4.5 years (range 3–7), patients had stable EDSS (29%), improved EDSS (57%) and only 1 (14%) progressed. Middle panel: whiskers box plots show EDSS modification according to the time from AHSCT. Right panel: no MRI activity was evidenced after AHSCT except for one patient who showed a new enhanced lesion 7 years after AHSCT, in absence of clinical relapse or clinical progression. (B) MSCs CXCL10 production after AHSCT. CXCL10 production was measured by ELISA in supernatants of MSCs isolated from seven patients before (pre) and after AHSCT (post) at three different passages, P2, P4, and P6, after LPS stimulation (24 h). MSCs POST AHSCT produce CXCL10 as MSCs PRE AHSCT. C) CXCL10 production in peripheral monocytes isolated from four MS patients before and after AHSCT. CXCL10 production was tested (Milliplex) in basal condition (−) and upon stimulation with LPS for 24 h (+LPS). As shown by column bar graph (pg/mL mean ± SEM), monocytes POST AHSCT produce CXCL10 at similar extent with respect to monocytes PRE AHSCT. MS, multiple sclerosis; MSCs, mesenchymal stem cells; AHSCT, autologous hematopoietic stem cell transplantation.

Figure 3

Signal transduction protein analysis. (A) The activation state of the signaling proteins p-38, CREB, JNK, and STAT-1 was investigated by Milliplex in five MS MSCs and five HD ones, on basal condition (−) and upon LPS stimulation (30 min for p-38, CREB and JNK, 2 h for STAT-1). For each analysis, 10 μg of total extracted proteins was used. The graphs show the ratio between phosphorylated protein and total protein levels (ph protein/tot protein), expressed as mean value ± SD. MS MSCs have higher level of ph p-38, ph CREB, ph JNK, and ph-STAT-1 than HD MSCs (*P < 0.05, Mann–Whitney test). (B) The activation state of p-38, CREB, JNK, and STAT-1 was also investigated in MS MSCs isolated after AHSCT (MS MSCs POST AHSCT) from the same patients analyzed in A) The phosphorylation level upon LPS stimulation is expressed as fold increase with respect to basal condition (ratio between ph protein/tot protein value in stimulated condition and ph protein/tot protein value in basal condition) and compared with pre-AHSCT (MS MSCs PRE AHSCT) and HD values (HD MSCs). Here, we show that after AHSCT MS MSCs preserve the same phosphorylation levels of p-38, CREB, JNK, and STAT-1 observed before AHSCT, and these are significantly increased compared with HD samples (*P < 0.04; **P < 0.001). HD, healthy donors; MSCs, mesenchymal stem cells; AHSCT, autologous hematopoietic stem cell transplantation; MS, multiple sclerosis; MSCs, mesenchymal stem cells.

Figure 4

Signaling protein expression profile. (A and B) STAT-1, NF-κB, and IκB expression was evaluated by real-time PCR in MSCs (A) and in peripheral monocytes (B) isolated from six MS patients (gray columns) and six HD (black columns). mRNA expression levels of STAT-1, NF-κB, and IκB were determined in basal condition (−) and upon LPS stimulation for 4 h (+LPS). mRNA expression is reported as ratio to β-actin (relative expression). Upon LPS stimulation, both MS MSCs (A) and MS monocytes (B) express higher levels of STAT-1, NF-κB, and IκB than HD ones (*P < 0.05, **P < 0.02, Mann–Whitney test). (C and D) Correlation between CXCL10 production (pg/mL) and STAT-1/NF-κB relative expression in MS MSCs (C) and MS monocytes (D) upon LPS stimulation. Each point in graphs represents an independent sample. Linear regression between CXCL10 (pg/mL) and STAT-1/NF-κB relative expression is statistically significant both in MSCs and in monocytes (P value and r² value are reported in each graph). (E) STAT-1, NF-κB, and IκB expression evaluated by real-time PCR in MS MSCs and peripheral monocytes after AHSCT and compared as fold increase with pre-AHSCT and HD values. Cells were isolated from the same MS patients before and after AHSCT. mRNA expression level of STAT-1, NF-κB and IκB was determined in basal condition (−) and upon LPS stimulation for 4 h (+LPS). Expression was evaluated as ratio to β-actin. Here, we show that transcription factors expression is still significantly increased in MS compared with HD (*P < 0.05) independently on AHSCT treatment. MSCs, mesenchymal stem cells; AHSCT, autologous hematopoietic stem cell transplantation.