Introduction
There is an important need for stromal cell lines that support neural cells and the mesenchymal stem cell (MSC) line SB623, transfected with the Notch-intracellular domain (NICD), appear to meet these criteria. In cultures of embryonic cortical neurons, SB623 cells produce extracellular matrix proteins which enhance and maintain neurite outgrowth [
1]. In neonatal hippocampal organotypic culture, SB623 cell-derived soluble trophic factors rescue neural cells subjected to oxygen-glucose deprivation [
2]. In experimental Parkinson's disease, grafting of SB623 cells efficiently reverses the degeneration of dopaminergic neurons by promoting endogeneous neuronal cell recovery [
3,
4]. And in stable stroke animal models, transplantation of SB623 cells reduces infarct size and promotes behavioral improvement [
5]. These studies validate one of the therapeutic applications of SB623 cells - to supply trophic factors for the endogenous neural cells after injury or disease.
Human marrow stromal cells are attractive for cell therapy because they can be obtained with minimal invasiveness and can be expanded in culture. However, as non-immortalized primary cells, MSCs have limited regenerative potential, committing to cellular senescence after extensive
ex vivo manipulation [
6,
7]. A potential upside of senescent cells is their robust cytokine secretome profile which could be beneficial in tissue regeneration. A potential downside is that the senescent-associated-secretome profile is thought to be pro-inflammatory [
8‐
10]. To date, intracerebral implantation of human SB623 cells in stroke-induced animals has not triggered any immunological adverse effect. Nevertheless, as SB623 cells are derived from MSCs that have undergone gene transfection and cell expansion in culture, we initiated the current study to determine whether SB623 cells display senescent-like properties. More importantly, we compare the immunomodulatory activity between SB623 cells and the corresponding parental MSCs. We demonstrate that SB623 cells, currently in a clinical trial for stable stroke (
http://clinicaltrials.gov/ct2/show/NCT01287936), retain the immunosuppressive activity of standard MSCs despite the appearance of a small number of senescent-like cells.
Materials and methods
Production of MSCs and SB623 cells
MSC and SB623 cells were produced as previously reported [
1,
2]. Briefly, human adult bone marrow aspirates (Lonza, Walkersville, MD) were plated in growth medium - αMEM (Mediatech, Manassas, VA) supplemented with 10% fetal bovine serum (FBS) (Hyclone, Logan, UT), 2 mM L-glutamine and penicillin/streptomycin (both from Invitrogen, Carlsbad, CA) for three days to obtain the marrow stromal cell (MSC) monolayer. After two passages, a portion of the culture was cryopreserved as MSCs. The remaining cells (passage 2) were transfected with the pCMV-h
NICD1-SV40- Neo
R plasmid using Fugene6 (Roche Diagnostics, Indianapolis, IN). After 7 days of selection with 100 μg/ml G418 (Invitrogen), the G418-resistant colonies were expanded and passed twice prior to cryopreservation as SB623 cells. This results in a uniformly transiently transfected population of MSCs.
qPCR and qRT-PCR
Two days after transfection with pN2-NICD plasmid, cells were lysed and DNA or RNA purified using Qiagen's QIAAmp DNA or RNeasy mini kits (Qiagen, Valencia, CA), correspondingly, according to the manufacturer's protocols. Quantitative real time PCR or RT-PCR analyses were conducted using QuantiTect Probe PCR or RT-PCR kits, respectively, on Lightcycler (Roche).
For exogenous-NICD (eNICD) qPCR analysis, purified RNA-free DNA samples were used at 65 ng (10000 diploid human genomes) per reaction and eNICD gene copy numbers were determined using eNICD-DNA-specific Taqman assay (forward primer: TTGGTCTTACTGACATCCACTTTG, reverse primer CAGACACTTTGAAGCCCTCAG, exo-NICD-specific probe [6-FAM]CCCAGTTCAATTACAGCTCTTAAGGCTAGAG[BHQ1a-6FAM])). Amplification signals were compared to those of pN2-NICD plasmid serially diluted in human genomic DNA (Clontech, Mountain View, CA); results expressed in numbers of plasmids per one human diploid genome (plasmids/cell). For expression analysis of a NICD target gene, human Hes1 and GAPDH (control) Taqman assays (Applied Biosystems, Carlsbad, CA) were used. Normalized Hes1 expression levels are presented relative to levels in non-transfected cells.
Phenotypic characterization by flow cytometry
For cell surface phenotyping, MSCs or SB623 cells were harvested with 0.25% Trypsin/EDTA (Invitrogen), washed in PBS/2% FBS, and re-suspended in 1 ml of PBS/2% FBS. Cells were then stained with fluorochrome-conjugated antibodies against CD29, CD31, CD34, CD44, CD45, CD73, CD90 (all from BD Biosciences, San Jose, CA) and CD105 (Invitrogen, Carlsbad, CA) for 15 minutes on ice. After one wash in PBS/2% FBS, cells were acquired using BD FACS Calibur. Analyses were done to assess the percentage of surface markers that are positive (CD29, CD44, CD73, CD90, and CD105) versus negative (CD31, CD34, and CD45) for mesenchymal cells using CellQuestPro program (BD Biosciences). To compare the density of specific surface molecule expression on MSCs versus SB623 cells, the delta mean fluorescent intensity (dMFI) was calculated - e.g., dMFI of CD44 = (MFI of CD44) - (MFI of IgG).
For intracellular protein detection of p16Ink4A and NICD, cells were fixed with 4% paraformaldehyde and permeabilized with PBS/0.1% TritonX-100. After two washes in PBS/2% FBS, cell pellets were resuspended in 200 ul of PBS/2% FBS and divided into two tubes, one for staining with phycoerythrin (PE)-conjugated IgG (control) and the other for staining with PE-conjugated p16Ink4A antibody (BD Bioscience) or PE-conjugated NICD antibody (eBioscience). For intracellular cytokine detection, cells were treated with BrefeldinA for six hours prior to harvest. After fixation and permeabilization, cells were incubated with fluorochrome-conjugated antibody against human GM-CSF (BD Bioscience), IL-1a (eBioscience, San Diego, CA), IL-6 (BD Bioscience), TGFβ1 (RnD Systems, Minneapolis, MN) for one hour followed by two washes in PBS/2% FBS. Acquisition and analysis of all samples were performed on BD FACS Calibur using CellQuestPro software.
Cell proliferation measurement
To quantify viable cell expansion, one million MSCs or SB623 cells were plated on Day 0 and cell counts by trypan blue exclusion were done on Day 3. For cell cycle profile after culture, one million MSCs or SB623 cells were fixed in 70% ethanol overnight at 4°C. After two washes in PBS/2% FBS, cells were incubated in one ml of staining buffer (50 μg/ml propidium iodide, 50 μg/ml RNAse) (Sigma, St. Louis, MO) in PBS/2% FBS for 30 min in the dark. Acquisition and analysis were done using CellQuestPro program on the FL-2 linear channel. For cell cycle kinetics over 5 days in culture, MSCs and SB623 cells were labeled with 5 μM of 5-(and-6)-carboxyfluorescein diacetate (CFSE) (Invitrogen) for 2 min at room temperature prior to culture. Flow cytometry acquisition and analysis were done on the FL-1 log channel.
Generation of monocyte-derived dendritic cells (Mono-DC)
Peripheral blood was obtained from healthy donors and mononuclear cells recovered from buffy coat preparations by Ficoll Paque (Amersham Pharmacia, Sweden) gradient separation. Mononuclear cells were re-suspended in RPMI/10%FBS and plated in a T-75 flask overnight. Non-adherent cells were discarded and the flasks were rinsed twice with PBS. Adherent monocytes were recovered using 0.25% trypsin/2 mM EDTA. Purity was assessed by staining with FITC-conjugated antibody against human CD14, a monocyte surface marker (Becton Dickinson) and was routinely shown to be > 90%.
For monocytic-to-dendritic cell differentiation assays, monocytes were cultured in RPMI-1640 (Mediatech) containing 10% FBS, 2 mM glutamine, 2 mM sodium pyruvate, 100 U/mL penicillin, 100 μg/mL streptomycin, 40 ng/mL granulocyte-macrophage colony stimulating factor (GM-CSF) and 20 ng/mL interleukin-4 (IL-4) (both from Peprotech, Rocky Hill, NJ) in the presence of MSCs or SB623 cells at a 10:1 monocyte to MSC or SB623 cell ratio. On Day 5, a subset of cultures were harvested by 0.25% trypsin/2 mM EDTA and stained with fluorochrome-conjugated antibodies against CD1a and CD14 (eBioscience). Data acquisition and analysis were done on the FACS Calibur using CellQuestPro software.
To assess the impact of MSCs and SB623 cells on the maturation of dendritic cells, monocyte-derived dendritic cells were generated in the presence of GM-CSF and IL-4. On Day 5, human TNF-α (10 ng/ml; Peprotech) was added to each well with or without MSCs or SB623 cells. As previous studies confirmed a role of cyclosporin A in hindering dendritic cell maturation [
11], addition of cyclosporin A (1 μg/ml; Santa Cruz Biotechnology, Santa Cruz, CA) in the absence of either MSCs or SB623 cells was included as an internal control. On Day 7, cells were incubated with a fluorochrome-conjugated monoclonal antibody against human CD86 (BD Bioscience), a co-stimulatory molecule for priming and activating naïve and memory T cells and analyzed on the BD FACS Calibur using CellQuestPro.
Ex vivo culture of human peripheral blood T cells
Human T cells were enriched from peripheral blood using the T-cell isolation kit (StemCell Technologies, Vancouver, Canada) according to the manufacturer's protocol. Enriched T cells were cultured in RPMI-1640/10% heat-inactivated FBS/pen/strep overnight prior to use. On Day -1, 10,000 MSCs or SB623 cells were plated per well of 96-well U-bottom plates. On Day 0 of the culture assay, 100,000 enriched T cells were transferred to each well with a pre-established MSC or SB623 cell monolayer. As an internal control, T cell cultures were maintained in the absence of MSCs or SB623 cells. On Day 7, a sub-optimal dose of 25 ng/ml of phorbol 12-myristate 13-acetate (PMA)/0.5 μM Ionomycin (both from Sigma-Aldrich) was added in the presence of BrefeldinA (eBioscience, 1:1000) for 6 hours prior to harvest for intracellular detection of interleukin-10 (IL-10) and interferon gamma (IFN-γ). For IL-17 producing TH17 cells, T cells were co-cultured with SB623 cells or MSCs in the presence of IL-23, or in the presence of IL-23 alone. After sub-optimal activation with PMA/Ionomycin in the presence of BrefeldinA, the cells were stained with fluorochrome-conjugated antibody against IL-17A (eBioscience) and analyzed by flow cytometry.
For regulatory T cell culture, human enriched T cells were co-cultured with MSCs or SB623 cells in the presence of human interleukin-2 (IL-2) (Peprotech, Rocky Hill, NJ) at a 10:1 T cells to MSC or SB623 cell ratio for 7 days followed by cell surface staining for CD4, a helper T cell marker and CD25, the IL-2 receptor alpha chain. For FoxP3 intracellular staining, cells were fixed and permeabilized with CytoFix/Perm (eBioscience). PE-conjugated antibody against FoxP3 (clone PCH101, eBioscience) was used at 1:50 dilution and flow cytometry analysis was done gating on lymphocytes. For assessment of constitutive IL-10 production, intracellular staining with fluorochrome conjugated antibody against IL-10 was performed without PMA/Io stimulation on Day 7.
Mixed lymphocyte reaction (MLR)
Human allogeneic mixed lymphocyte reaction was established using peripheral blood from unrelated healthy volunteers. To obtain responder cells, T cell enrichment using a commercial T-cell rosette separation kit (Stem Cell Technologies) was done based on the manufacturer's protocol. Enriched T cells (= responders) were labeled with 5 μM of 5-(and-6)-carboxyfluorescein diacetate (CFSE) (Invitrogen) for 2 min at room temperature. CFSE-labeled lymphocytes were then plated in a 96-well U bottom plate at a concentration of 100,000 cells per 100 μl per well. To obtain stimulator cells, peripheral blood buffy coat mononuclear cells were recovered after Ficoll-density gradient centrifugation and red blood cell lysis buffer (Sigma-Aldrich) was added for 10 min at 37°C. 100,000 stimulator cells were added to a tube containing 10,000 MSCs or SB623 cells; and the mixed cells were then centrifuged and re-suspended at 110,000 mixed cells per 100 μl. 100 μl of stimulator/MSC cell mix or 100 μl of stimulator/SB623 cell mix was added to each well of CFSE-responder cells. To assess the activation state of T cells in the MLR, cells were harvested on Day 2 and stained with a fluorochrome conjugated antibody against CD69 (BD Bioscience), an antigen induced on activated T cells. To monitor cell proliferation kinetics of T cells in the MLR, cells were harvested on Day 7 and stained with PE-conjugated CD4 antibody (BD Bioscience). Flow cytometry data acquisition was done on BD FACS Calibur, gating on CD4+ lymphocyte gate, and analysis was done using CellQuestPro.
Xenogeneic MLRs were established using postnatal day 9 Sprague-Dawley rat glial mix cells as stimulators and human peripheral blood T cells as responders. Briefly, rat brains were harvested and triturated prior to treatment with 0.25% trypsin (Invitrogen) for 30 min. Cell suspensions were filtered through a 70 μM cell strainer and overlaid on Ficoll prior to density centrifugation. Glial mix cells were cultured in DMEM/F12 (Mediatech)/10%FBS/pen-strep for 14 days prior to use in the MLR. Xenogeneic MLRs were performed at a similar cell ratio as allogeneic MLRs (100,000 glial mix: 100,000 CFSE-labeled human T cells: 10,000 MSCs or SB623 cells) over a 5-day period. CFSE dilution of human CD3-gated T cells was assessed by flow cytometry.
Statistics
Statistical assessments (SigmaStat, Systat Software, Chicago, IL) were made for SB623 or MSC groups to determine if there were differences either between those two groups or in some cases compared to the internal assay controls. To compare co-cultures with SB623 cells to those with MSCs (n = 3-6 matched lots), Tukey's pairwise comparisons were made. To compare co-cultures with SB623 cells or MSCs to internal experimental controls (when n>1), a general linear model ANOVA, followed by Tukey's pairwise comparisons was used. An alpha value of 0.05 was used to assess if the means were significantly different. Data are reported as mean ± standard deviation.
Discussion
Ex vivo manipulation of MSCs has been shown to induce their cellular senescence [
6,
7] and senescent cells have been described to have a pro-inflammatory secretome [
8‐
10]. Because SB623 cells are derived from
ex vivo manipulated MSCs, we investigated the possible senescent onset in multiple lots of SB623 cells and compared the immunomodulatory activity of SB623 cells to that of the parental MSCs.
Morphologically, SB623 cells resemble their parental MSCs. Phenotypically, SB623 cells expressed all the standard MSC surface markers (CD90, CD105, CD29, CD44, CD73), although there was an increased density of CD44 and CD73. A small number of SB623 cells expressed surface CD54, an inter-cellular adhesion molecule serving as the ligand for LFA-1, the lymphocyte function-associated antigen. SB623 cells displayed a similar cytokine expression profile as parental MSCs. The effect of Notch in MSCs has been under much investigation. One study has implicated a function of Notch in promoting cellular senescence of rodent cells [
35]. In our system, we transiently expressed
NICD in human MSCs by DNA plasmid transfection. Analyzing for two senescent markers - beta-galactosidase positivity and p16Ink4A expression, we noted a small number of senescent-like cells within each lot of SB623 cells. From cell cycle profile and kinetics, we observed reduced proliferation in SB623 cultures compared to the parental MSCs. This reduced proliferation is most likely not mediated by exogenous
NICD transient expression as we observed similar reduction in growth for MSCs transfected with an empty expression vector (data not shown). Therefore, we suspect that the small number of senescent-like cells within each lot of SB623 cells is a reflection of the extended time in culture (~2 months).
As noted above, some studies have highlighted the pro-inflammatory secretome of senescent cells [
8‐
10]. To date, we did not observe immunological side effects from SB623 cell implantation in rats. Nevertheless, as we noted a small population of senescent-like cells within each lot of SB623 cells, we initiated various cellular immune assays to compare their immunomodulatory activity to parental MSCs in more detail. In an allogeneic mixed lymphocyte reaction (MLR), we demonstrated that similar to MSCs, SB623 cells attenuated the activation of CD4+ T cells as evident by reduction in CD69 (an early T cell activation marker) and HLA-DR (an activation marker on both T cells and monocytes). In experimental rodent stroke, intracerebral implantation of SB623 cells elicits functional recovery [
5]. As the glial cells are among the common antigen presenting cells in the nervous system, we assessed the efficiency of SB623 cells in suppressing the proliferation of human T cells stimulated by rat glial mix cells. We demonstrate that SB623 cells elicited immunosuppressive activity in this xenogeneic MLR assay, comparable to parental MSCs.
In the context of immune modulation, MSCs have been reported to impact both the innate and acquired immune cells [
25]. In a standard mono-dendritic cell differentiation assay with GM-CSF and IL-4, we demonstrate that the inclusion of SB623 cells in the mono-dendritic cell differentiation culture reduced the production of dendritic cells (CD1a+CD14-) to similar extent as the parental MSCs. In a 2-day TNF- α mediated dendritic cell maturation assay, the inclusion of SB623 cells reduced the density of CD86 co-stimulatory molecules, as measured by mean fluorescent intensity. Interestingly, the reduction in CD86 surface expression was significantly higher in the presence of SB623 cells than the parental MSCs. A recent study reported that activated MSCs secrete soluble TNF-α receptors which, in turn, attenuate systemic inflammation [
36]. As TNF- α is commonly used to induce dendritic cell maturation, we hypothesize that a differential expression and/or secretion of TNF- α receptors between SB623 cells and MSCs could explain our current observations. Additional studies are warranted to address this possible underlying mechanism of action.
To assess the impact of SB623 cells on the acquired immune cells, we performed co-cultures of human enriched T cells with SB623 cells or MSCs and assessed the activated T cell secretome profile following stimulation with sub-optimal dosage of PMA/Ionomycin. By intracellular staining with antibodies against IL-10 and IFN-γ, we observed a robust skewing in the activated immune secretome profile with more than 95% of cells expressing IL-10 and less than 5% expressing IFN-γ. The detection of predominantly IL-10 expressing cells is in line with previous reports that MSCs promote the anti-inflammatory secretome of T cells. The lack of IFN-γ production was unexpected since Th1 cells are known to produce both IFN-γ and IL-10, not just IL-10 alone. However, IL-10 has been shown to downregulate IFN-γ production [
37‐
39]. It is therefore possible that in the presence of SB623 cells or MSCs, the level of IL-10 produced was high enough to form a negative feedback loop on the production of IFN-γ. A few studies have highlighted the production of IL-10 from Th1 T cells as a "self-control" mechanism [
40,
41] and Notch activation has been associated with this process [
42]. TGFβ1 and IL-6, both of which are known to be produced by MSCs, have a role in Th17- and Th1- T cell development. To determine if SB623 cells impact the number of Th17-T cells, we performed co-cultures of T cells with SB623 cells or MSCs in the presence or absence of IL-23, a cytokine supportive of human Th17-T cell development. By intracellular staining with an antibody against IL-17A, we detected an average of less than 1.5% Th17-T cells, with no significant differences between co-cultures with SB623 cells or the parental MSCs. While it is surprising to see a positive impact of marrow stromal cells on Th17 cell number in culture, one study to date has highlighted this property of fetal marrow stromal cells [
43]. It is also important to note that the percentage of IL-17-producing T cells is relatively low compared to the percentage of IL-10 producing T cells and thus, may explain why transplantation of MSC elicits an overall immunosuppressive outcome [
24,
25].
Another immunomdulatory property of TGFβ1 and Notch ligands is in the regulation of regulatory T cells (Tregs) [
13,
16,
21]. As both TGFβ and Notch ligand are expressed by SB623 cells and the parental MSCs, we next investigated the impact of SB623 cells on T cells cultured in the presence of IL-2, a cytokine important in Treg cell development. In the presence of SB623 cells, we observed a higher percentage of cells expressing surface CD25, the IL-2 receptor alpha chain. Although CD25 is commonly used to identify Tregs [
22,
23], this surface marker is also present on activated T cells [
44]. An alternate common marker for Tregs is the forkhead family transcription factor FoxP3[
45]. By intracellular staining with an antibody against FoxP3, we demonstrated a higher percentage of FoxP3+ cells when SB623 cells or MSCs were included. As FoxP3 is not exclusively expressed in Tregs [
46], we measured the intracellular expression of IL-10, a cytokine constitutively produced at low levels by Tregs [
47,
48]. We consistently detected a small but higher percentage of CD4+ T cells expressing IL-10 in cultures with SB623 cells or MSCs compared to the controls. These results suggest that SB623 cells, like the parental MSCs, may have a role in supporting T cells having various features of regulatory T cells.
SB623 cells are derived from NICD-transfected MSCs and expanded in the absence of exogenous cytokines. As such, SB623 cells retained the standard phenotypes and morphology of conventional MSCs. Compared to the early passage MSCs, SB623 cells contained a small number of senescent-like cells as a result of cell expansion in vitro. Nevertheless, we show here that SB623 cells effectively suppressed T cell proliferation in MLR and modulated the T cell secretome profile as efficiently as the parental MSCs. The current study demonstrate that the transient overexpression of exogenous NICD with subsequent expansion of human MSCs preserved, and in some cases increased, their immunosuppressive activity.
Authors' contributions
MD conceived and designed the study, performed immunoassays and flow cytometry, analyzed and interpreted data, and wrote the manuscript. CCT participated in data analysis and interpretation, performed statistical analysis, and edited the manuscript. IA prepared DNA and RNA samples, performed PCR, analyzed and interpreted PCR data. MM designed and coordinated the production of MSCs and SB623 cells. CCC directed the study, designed PCR primers for the detection of exogenous NICD, and helped with data interpretation. All authors read and approved the final manuscript.