Monocyte-derived cytokines in multiple sclerosis

Abstract

MS is an inflammatory, presumably autoimmune, disease mediated by the activation of T cells, B cells and monocytes (MO). Inflammation is thought to occur early during the relapsing-remitting phase of MS (RRMS), whereas in the later phases of MS such as secondary progressive MS (SPMS), inflammation tends to diminish. Our objective was to compare the types and amounts of proinflammatory and regulatory cytokines produced by MO from relapsing–remitting patients with or without treatment with IFN-β (RRMS⁺ therapy, RRMS⁻ therapy), respectively, from secondary progressive patients (SPMS) and from healthy controls (HC). MO were isolated by a density-gradient technique and three different techniques (RNase protection assay, ELISA and intracellular cytokine staining) were used to assess cytokine levels. An increase in IL6, IL12 and TNF-α was observed by all three methods for RRMS⁻ therapy and for SPMS patients compared to HC and RRMS⁺ therapy patients. We conclude that proinflammatory and regulatory monokines can be derived from MO of MS patients and that these levels are modulated by IFN-β therapy. Although it is believed that inflammation tends to diminish in SPMS patients, our data show that inflammatory cytokines continue to be released at high levels, suggesting that IFN-β or IL10 treatment may be beneficial for this group.

Introduction

The most consistent pathological finding in multiple sclerosis (MS) is the presence of perivenular T cell and macrophage infiltrates among the multifocal areas of myelin breakdown and gliosis in the central nervous system (CNS). The T cells are often activated, proliferating and releasing cytokines, whereas ultra-structurally, macrophages (MΦ) have been seen literally engulfing myelin. Although the aetiopathogenesis of MS is still unknown, overwhelming evidence indicates that autoimmune mechanisms play a large role. T cell–MΦ interactions are often linked to the induction of an autoimmune inflammatory process. As MS progresses inflammation tends to diminish, with many lesions consisting mainly of MΦ/microglia [1,2] and blood monocytes (MO) [3].

Cytokines have been implicated widely in the immunopathogenesis of MS, but the exact role of cytokines and how they are regulated in this disease remains controversial [4,5]. Cytokines are produced by activated T cells, MO, MΦ and microglial cells, all of which are found in MS lesions. Inflammatory and regulatory cytokines produced by MO include IL-1, IL, IL-6, TNF-α, IL-10 and IL-12 [6]. Th1 cytokines including IL-12 produced by MO are elevated in MS patients and these increased levels correlate with disease severity [3, 7, 32]. In our in vivo longitudinal study of IFN-β treated patients, CD40L expression was reduced on T cells and MO, as was CD86 expression on MO, after 3 months of therapy. Similar findings were observed by others [3] with respect to the expression of IL-12 [8–10]. Liu et al. [11] demonstrated that IL-10 mRNA levels were increased in MS patients. We concur with this observation for mRNA levels; however, the secreted levels of IL-10 are not increased in MS patients [3,32]. In addition, IFN-β decreased the expression of CD86 and CD40 on MO while it increased expression of IL-10. We and others have published that IL-10 has a profound down-regulating effects on CD86 not only in MS [12,32] but also in HIV patients as well as normal individuals [13,14].

IFN-β is the first cytokine used in therapy shown to alter the natural course of RRMS [15–18], but its mechanism of action is not well understood. We hypothesized that IFN-β might alter the cytokine response of MO. In this study, we performed a thorough, semiquantitative analysis of transcription and translation of IL-1β, IL-1α, IL-6, TNF-α, IL-10 and IL-12 in MS patients at various stages of disease progression and/or treatment. The results indicate increased levels of monocytic IL-1β, IL-6, IL-12 and TNF-α production in MS patients as compared to HC levels.

Materials And Methods

Patients

Blood samples (50 ml of heparinized blood) were obtained from patients attending the multiple sclerosis clinic at the Ottawa Hospital General campus and from healthy age- and sex-matched laboratory volunteers. The study was approved by the Human Research Ethics Committee, Ottawa Hospital and all donors provided written consent. Table 1 summarizes donor information. MS patients had clinically definite MS [19] and each patient's expanded disability status scale (EDSS) score was determined at each visit [20].

Table 1.

Patient description

	Patient information
				EDSS score
	Number	Female/male ratio	Age	<3	>3
HC	9	5 : 4	Median: 30	NA	NA
			Range: 22–47
RRMS− therapy	10	7 : 3	Median: 40	n = 4	n = 5
			Range: 18–54	Mean = 2	Mean = 5·5
RRMS+ therapy*	6	4 : 2	Median: 41	n = 5	n = 1
			Range: 29–46	Mean = 4	Mean = 1·7
SPMS	10	7 : 3	Median: 46		n = 9
			Range: 25–69		Mean = 6

^*Patients on IFN-β therapy for at least 3 months. Patients were seen in the multiple sclerosis clinic at the Ottawa Hospital General Campus. Informed written consent was obtained from each patient.

MO isolation and culture conditions

MO were isolated as described previously employing the Optiprep method [21]. The isolations yielded cell preparations containing greater than 95% pure MO as determined by CD14, HLA-DR and CD4 expression and fewer than 1% CD3⁺ T cells. Cells were cultured for 24 h in polypropylene tubes at a cell density of 10⁶ cells/ml. Because MO adhere strongly to plastic surfaces, polypropylene tubes were used to minimize the adherence of these cells to the walls of the culture tubes. Several MO cultures were established from a MO separation performed for each patient and HC. The cultures were established in RPMI, 10% FCS and 1% of penicillin and streptomycin (Life Technologies, Toronto, ON, Canada). Some cultures were treated with rhIL-10 (0·1 or 1 ng/ml) (Schering Plough Research Institute Kenningworth, NJ, USA) or IFN-β (10 or 100 units) (Ares-Serono, Montreal, PQ, Canada). For each MO culture condition one culture was used for intracellular cytokine staining, whereas the second culture was used to measure the level of secretion of cytokines and the amount of mRNA of each cytokine. In additional cultures, 100 ng/ml of lipopolysaccaharride (Sigma, St Louis, MO, USA) were added to the cultures and all three cytokine detection assays were performed.

Sample preparation for flow cytometry analysis

The staining of freshly isolated and cultured MO was performed as described previously [22]. Flow cytometry analysis included autofluorescence control (tube 1), colour compensation controls (FITC, PE and PE-Cy5) (tubes 2–4), isotype controls and combinations of antibodies to CD45 CD3, CD8 (Caltag, Burlingame, CA, USA), CD4, HLA-DR and CD14 (Sigma). The mean fluorescent intensity (MFI) of the isotype controls were identical to those observed for autofluorescence.

Intracellular staining of cytokines in MO

MO were isolated and cultured as described above. Monensin (GolgiStop, 4 µl/6 × 10⁶ cells; Pharmingen, San Diego, CA, USA) was added in the final 7–8 h of culture to inhibit Golgi transport and to induce accumulation of cytokines in the Golgi body. Cells were harvested and washed with PBS (Sigma, St Louis, MO, USA), supplemented with 0·5% FCS and 0·1% NaN₃ (BDH Inc., Toronto, ON, Canada). All cells were treated with heat-aggregated human IgG (1 mg/ml) for 10 min followed by extracellular staining as described above. The cells were washed, incubated again with heat-aggregated human IgG (1 mg/ml) for 10 min to block Fc receptor binding, fixed and permeabilized for 15 min with paraformaldehyde (PFA) (BDH Inc.), and 0·1% saponin (Sigma). Permeabilized cells were washed with PBS supplemented with 0·5%FCS, 0·1% NaN₃ and 0·1% saponin. Cells were aliquoted and incubated with PE-conjugated monoclonal antibodies for IL-10 (Biosource, Camarillo, CA, USA), IFN-γ (Caltag), IL-6, IL-10, IL-12 or TNF-α (Pharmingen) for 25 min. The monoclonal antibody specific for IL-12 recognized both human IL-12p70 and p40 subunits. Cells were washed with PBS supplemented with 0·5% FCS, 0·1%NaN₃. Cells were analysed on an Epics MCL flow cytometer. Mean fluorescent intensity (MFI) for each of the autofluorescent control groups and the specific cytokines being detected were determined.

The isotype control used in this study was the anti-IL-2 antibody conjugated to PE. It was used to determine non-specific binding to MO and to control for contaminating T cells. The anti-IL-2 antibody did not stain fresh or cultured MO and the levels of contaminating CD4⁺/IL2⁺ T cells were less than 1%. The cells were analysed using an Epics XL MCL flow cytometer and the analysis and re-analysis of the listmode files were performed using the Epics software (Beckman-Coulter, Hialeah, FL, USA), or WinMdi (Scripps Foundation, San Diego, CA, USA).

The percentage and MFI of cytokine-expressing MO subpopulations were tabulated for each sample. MFI values were used to compare the amounts of cell surface marker expression. Flow cytometry histograms were superimposed and the observation of a cell population shift of greater than 2× in log fluorescence intensity on the MFI log scale was regarded as significant.

Total RNA extraction and Riboquant RNase protection assay

Total RNA was isolated from harvested MO using Tri-Reagent (Boehringer Mannheim, Indianapolis, IN, USA). RNA purity was assessed by agarose gel electrophoresis (1·8%). Equal quantities of total RNA (1–5 µg) were used for cytokine mRNA detection using the Riboquant RNase Protection Assay (Pharmingen) as per the manufacturer's instructions. ³²P-labelled probes specific for the hCK-2 template (Pharmingen) were synthesized. The hCK-2 template contains probes specific for IL-12 p35, IL-12p40, IL-10, IL1-β, IL-1α, IL-1Ra, IL-6, IFN-γ and the housekeeping genes L32 and GAPDH. Following probe synthesis, RNA was extracted and RNA-probe hybridization was performed for 12–16 h at 56°C. Hybridized RNA samples were subjected to RNase treatment, followed by purification of protected probe-mRNA fragments. Protected fragments were separated via electrophoresis on a denaturing 5% polyacrylamide gel and were detected by phosphor imagery (Scancontrol Software, Storm Phosphoimagery Systems, Molecular Dynamics, San Diego, CA, USA). Bands were analysed and quantified using ImageQuant Software (Molecular Dynamics). Housekeeping genes incorporated into the probe template allowed for direct comparison of cytokine mRNA levels between multiple samples and gels.

Cytokine Elisa

ELISA kits for IL1-β, IL-6, IL-10, IL-12 (p70) and TNF-α were obtained from Bio-Source Int. (Camarillo, CA, USA) and the assays were performed as instructed by the manufacturer. The ELISA plates were read in a Spectra Count ELISA plate reader (Canberra Packard, Meridien, CT, USA). The cytokine levels in each sample were calculated from the standard curve generated using positive and negative control reagents from the kits.

Statistical analysis

Statistical analysis was performed using GraphPad software, StatMate version 1. Data were analysed using a one-way anova, and if significant Newmann–Keuls post-test for multiple comparisons was used.

Results

Basal levels of cytokine expression in MO from MS patients

MO enter the circulation from the bone marrow, and within 30 h home to different tissues of the host in response to endogenous or exogenous stimuli [23]. Given the short half-life of circulating MO, the activation of these cells in MS patients should occur shortly after leaving the bone marrow and before migrating into tissues. The objective of the first series of experiments was to ascertain if there is a difference in the transcriptional and translational control of cytokine production in MO from untreated and IFN-β-treated RRMS and from SPMS patients compared to HC.

As seen in Fig. 1, the mRNA levels for IL-12p40, IL-10, IL-1α, IL-1β, IL-1Ra and IL-6 expression are significantly elevated or have a tendency to be higher in SPMS patients compared to HC. Figure 2 summarizes normalized density measurements (band volumes) of the RNA protected bands from all of the patients. The band volumes were normalized by taking into consideration the housekeeping genes (L32 and GAPDH) and the background density for each autoradiograph. After this was performed a final band volume or RNASE protection assay value (RPA value) was calculated. As can be observed, IL-12p40, IL-10, IL-1α, IL-1β, IL-1Ra and IL-6 are found in lower levels in RRMS⁻ therapy patients and are often further reduced in RRMS⁺ therapy patients compared to SPMS patients. IFN-γ was also detected but no significant differences were observed between the RRMS⁻ therapy and SPMS patients, although RRMS⁻ therapy patients had lower IFN-γ levels than other patients in the study (results not shown).

Fig. 1.

Cytokine mRNA levels of MO from MS patients. MO were isolated from peripheral blood and the basal cytokine mRNA levels after 24 h of culture were measured using an RNA protection assay. The figure is representative of the data that were obtained for five experiments. The first and second columns are the positive and negative controls, respectively, provided by the kit. The image was obtained using a phosphorimager.

Fig. 2.

Cytokine mRNA levels of MO from MS patients with or without culture treatment with IFN-β or rhIL-10. MO were isolated from peripheral blood from HC or MS patients and were treated with medium, IFN-β (100 IU/ml) or rhIL-10 (1 ng/ml) for 24 h. The cytokine mRNA levels of the cytokines were measured using an RNA protection assay. Data was obtained from HC (n = 9), RRMS⁻ therapy (n = 10), RRMS⁺ therapy (n = 6) and SPMS (n = 10). The mean and standard deviation of the data are shown.

Intracellular cytokine protein levels were determined for IL-1β, IL-6, IL-10, IL-12 and TNF-α. Figure 3 shows a representative result from gated MO. TNF-α was used as a positive control. There was only a tendency for IL-1β, IL-6 and TNF-α to be increased in MS patients compared to HC, and SPMS patients had higher levels than RRMS patients regardless of treatment status (results not shown)

Fig. 3.

Intracellular cytokine assay. MO from HC or patients were cultured for 24 h in the presence or absence of IFN-β (100 IU/ml) or rhIL-10 (1 ng/ml). Representative data from an SPMS MO cultured in medium are shown. Data were obtained from HC (n = 9), RRMS^-therapy (n = 10), RRMS⁺ therapy (n = 6) and SPMS (n = 10). The isotype control used was the anti-IL2 antibody from Pharmingen.

The secreted levels (Fig. 4) of IL-1β, IL-6, IL-10, IL-12 and TNF-α followed a very similar pattern to that observed with mRNA levels (Fig. 2). Significantly, elevated levels of most cytokines examined were observed in SPMS patients compared to both untreated and IFN-β-treated RRMS patients and HC. These cytokines were usually found in decreasing levels in RRMS⁻ therapy and RRMS⁺ therapy patients. HC did not produce significant levels of IL-1β, IL-6 or TNF-α but MO from all the MS patients examined usually did, even though these cells are newly formed and were not stimulated in vitro. The SPMS patients produced 10-fold higher amounts of IL-12 than that observed for untreated RRMS and RRMS receiving IFN-β therapy. All groups produced similar levels of IL-10 (results not shown).

Fig. 4.

Secreted levels of cytokines of MO from MS patients with or without treatment with IFN-β or rhIL-10. MO were isolated from peripheral blood from HC or MS patients and were treated with medium, IFN-β (100 IU/ml) or rhIL-10 (1 ng/ml) for 24 h. The secreted levels of the cytokines were measured by ELISA. Data were obtained from HC (n = 9), RRMS⁻ therapy (n = 10), RRMS⁺ therapy (n = 6) and SPMS (n = 10) patients. The mean and standard deviation of the data are shown.

Effect of exogenous IFN-β and rhIL10 on monokine expression in MS patients

IFN-β therapy induces profound changes in the type and in the amount of cytokines produced by MO in MS. The response of MO from MS patients undergoing IFN-β therapy to further in vitro stimulation with IFN-β or with rhIL-10 was assessed to determine if these MO were refractory to further suppressive in vitro effects of IFN-β or rhIL10.

MO from MS patients and from HC were cultured with a suboptimal concentration of IFN-β (100 IU/ml) or rhIL-10 (1 ng/ml). Suboptimal amounts were used because if the concentration of exogenous cytokine used is too high, it may mask subtle effects that may be observed and may reduce the sensitivity of the experiments, especially in the cells from RRMS⁺ therapy patients. Therefore, we chose an IFN-β concentration of 10 and 100 IU/ml. The responses of MO from all MS patient categories and HC to LPS (100 ng/ml) were also assessed.

mRNA levels for IL-6, IL-1Ra, IL-Iβ, IL-1α, IL-12p40 and IL-10 were detected by a RNase protection assay showing that MO from RRMS⁺ therapy patients did not respond to any further treatment with IFN-β (Fig. 2). IFN-β reduced the expression of the mRNA levels in other patient categories. The lack of suppression by IFN-β at 100 IU/ml in these cultures may be due solely to the low levels of IFN-β used in these experiments. rhIL-10 (1 ng/ml) did reduce further the mRNA levels of IL-6 and IL-1Ra in RRMS⁺ therapy patients. rhIL-10, however, reduced the mRNa levels of all of the cytokines in RRMS not treated and SPMS patients.

MO from each control and patient group were stimulated in vitro with IFN-β or rhIL-10. The intracellular cytokine protein levels of IL-1β, IL-6, IL-10, IL-12 and TNF-α were determined. No significant changes were observed compared to baseline levels between experimental in vitro stimulations in all patient and control categories (results not shown)

The secreted levels of IL-1β, IL-6, IL-12 and TNF-α followed the same trend that was observed for mRNA levels (Fig. 4). MO from all groups of patients, but not HC, secreted significant levels of IL-1β, IL-6 and TNF-α within 24 h. The levels of these cytokines did not differ between RRMS⁺ therapy and RRMS⁻ therapy patients. SPMS patients produced approximately 10-fold more IL-12 than that observed for RRMS⁻ therapy or RRMS⁺ therapy patients. Neither IFN-β nor rhIL-10 affected the levels of IL-1β or IL-12 secreted by MO from RRMS⁺ therapy patients. However, in vitro stimulation with IFN-β up-regulated IL-6 and TNF-α secretion by MO from IFN-β -treated RRMS patients (Fig. 4). In contrast, in vitro IFN-β stimulation of MO from untreated RRMS patients resulted in reduced TNF-α secretion (Fig. 4), with no statistically significant effect on the levels of IL-1β, IL-6 and IL-12 secretion. Interestingly, in vitro stimulation of MO from SPMS patients with IFN-β reduced secretion of both TNF-α and IL-6 while having only a minimal effect on IL-1β and IL-12 secretion (Fig. 4).

The in vitro effects of rhIL-10 on monokine secretion were much more similar in all patient groups and in controls. rhIL-10 significantly suppressed the levels of IL-6, IL-1β, IL-12 and TNF-α in MO from all MS patients and from HC. These results demonstrate that RRMS patients can respond further to in vitro stimulation with rhIL-10 or IFN-β, although the responses might be different depending on the clinical form of MS and whether the patient is receiving treatment.

The LPS response of MO from all MS patients and from HC as assessed in the RNase protection assay, and in assays measuring intracellular cytokine levels and secreted cytokine levels, were not significantly different, demonstrating that MO remain similarly responsive to stimuli and to the same extent (results not shown).

Discussion

The purpose of this investigation was to determine the effects of MS disease on specific cytokine levels produced by MO. The results were somewhat surprising, as MO from MS patients are activated in contrast to HC. MO are derived from myeloid precursors from the bone marrow and their half-life in the peripheral circulation is approximately 30 h [23]. Within this time frame these cells are being activated before they enter the tissues, where they differentiate into tissue-specific macrophages, microglia, dendritic cells, etc. [24,25].

Enriched, non-activated MO (Fig. 1) were used to test the levels of monocytic cytokines from MS patients (RRMS⁺ therapy or RRMS⁻ therapy and SPMS) and from HC. Isolation of MO from HC and from MS patients was performed on the same day. MO were not activated by the isolation procedure or by culture conditions, as proinflammatory cytokines were not induced in HC samples. LPS was able to induce the proinflammatory cytokines in both HC and in all MS patient categories within 24 h of culturing, but no significant statistical differences in monocytic cytokine levels after LPS stimulation were observed between HC and MS patient categories. These data support our observation that MS patients have activated and responsive MO.

The mRNA levels and secreted cytokine levels IL-1β, IL-6, IL-12 and TNF-α are concordant except for the IL-10 data. The most significant increases in mRNA levels for the cytokine IL-1β, IL-6, IL-12p40 and IL-10 were observed in SPMS patients compared to controls. However, the large standard deviation suggests that there may be subgroups within the SPMS population. This is currently being investigated.

Intracellular staining for cytokine production as determined by flow cytometry showed no differences between HC and MS samples (data not shown); however, ELISA data revealed extremely high quantities of cytokines being released from MS patient MO compared to MO from HC. This discrepancy in data could be due to the time-course of cytokine release in relation to when monensin, a Golgi inhibitor, was added to the samples. In addition, mRNA decay within eukaryotic cells is not a default process. It seems there is no indiscriminate degradation of mRNA but is performed by a precise process dependent on a variety of cis-acting elements and trans-acting factors [26]. This would enable the cell to accumulate mRNA species, translate others and accumulate protein within the cell. The precise mechanism controlling degradation of cytokine mRNA within MO is not known. However, this process would affect the overall production and release of cytokines in the microenvironment significantly.

MS pathogenesis is very complex and can no longer be explained solely by T cell activation. Inflammation is believed to be an initial feature of MS followed by demyelination [27]. Inflammatory CNS infiltrates, as well as the production of pro- and anti-inflammatory cytokines and chemokines, all have proposed roles in MS disease pathogenesis and severity [28,29]. However, γδ T cells may be effector cells in MS and contribute to inflammation via the production of proinflammatory or regulatory cytokines [30]. Collectively, our experiments reveal that MS disease correlates with increased monocytic proinflammatory/regulatory cytokine levels, specifically IL-1β, IL-6 and TNF-α (Fig. 2).

Cytokine up-regulation by MS patient MO makes these cells important contributors to inflammation, Th1 regulation and MS disease activity. Of interest, there are the increased levels of IL-1β and IL-6 produced by RRMS therapy patients. These inflammatory and regulatory cytokine levels are usually decreased in patients undergoing IFN-β therapy and are comparable to HC levels. These therapeutic agents may act on the immune system by depressing the levels of these cytokines, thus improving disease in MS patients.

MO from all MS patients, even those on IFN-β therapies, remain responsive to further treatment with cytokines such as IL-10. This cytokine depressed further inflammatory and regulatory monocytic cytokines and its own production within 24 h of culture initiation. Both the transcription and translation levels of the monocytic cytokines were depressed by rhIL-10. rhIL-10 was able to depress the monocytic cytokine levels of RRMS⁻ therapy and SPMS patients. The lack of statistical significance for RRMS⁺ therapy is due probably to the fact the monocytic cytokine levels are already low. However, IFN-β treatment at 10 or 100 IU/ml was not effective in depressing the monocytic cytokine levels in all MS patient categories, even in patients being treated with IFN-β. Our observation that IFN-β in vitro stimulation reduced secretion of IL-6 and TNF-α by MO derived from SPMS patients suggests a potential mechanism of this therapeutic agent in SPMS patients, because it has been demonstrated that IFN-β treatment is also effective in this clinical subgroup of MS patients [31]. On the other hand, IFN-β in vitro stimulation of MO obtained from IFN-β -treated RRMS patients resulted in up-regulation of proinflammatory cytokines IL-6 and TNF-α, suggesting that the cytokine interplay is very complex and that the delicate balance between different cytokines is more important than just a simple up-regulation of pro- or anti-inflammatory cytokines.

This study sheds some light on the role of MO in MS immunopathogenesis. T cells have long been thought to be the main effector cell in MS responsible for inflammatory and regulatory cytokine production. This study shows that MO play an important role in MS immunological and inflammatory processes. Increased levels of IL-1β, IL-6 and TNF-α produced by MS patients’ MO might contribute to increased cellular activation and proliferation, tissue damage, altered blood–brain barrier permeability and, thus, the development of MS disease. The exact trigger responsible for up-regulating monocytic inflammatory cytokine transcription and/or translation has yet to be elucidated. Thus, further study into monokine regulation in MS is warranted to expand firmly on a role for MO in this disease.