Do cryopreserved mesenchymal stromal cells display impaired immunomodulatory and therapeutic properties?

Abstract

We have recently reported that therapeutic mesenchymal stromal cells (MSCs) have low engraftment and trigger the instant blood mediated inflammatory reaction (IBMIR) after systemic delivery to patients, resulting in compromised cell function. In order to optimize the product, we compared the immunomodulatory, blood regulatory, and therapeutic properties of freeze-thawed and freshly harvested cells. We found that freeze-thawed MSCs, as opposed to cells harvested from continuous cultures, have impaired immunomodulatory and blood regulatory properties. Freeze-thawed MSCs demonstrated reduced responsiveness to proinflammatory stimuli, an impaired production of anti-inflammatory mediators, increased triggering of the IBMIR, and a strong activation of the complement cascade compared to fresh cells. This resulted in twice the efficiency in lysis of thawed MSCs after 1 hour of serum exposure. We found a 50% and 80% reduction in viable cells with freshly detached as opposed to thawed in vitro cells, indicating a small benefit for fresh cells. In evaluation of clinical response, we report a trend that fresh cells, and cells of low passage, demonstrate improved clinical outcome. Patients treated with freshly harvested cells in low passage had a 100% response rate, twice the response rate of 50% observed in a comparable group of patients treated with freeze-thawed cells at higher passage. We conclude that cryobanked MSCs have reduced immunomodulatory and blood regulatory properties directly after thawing, resulting in faster complement-mediated elimination after blood exposure. These changes seem to be paired by differences in therapeutic efficacy in treatment of immune ailments after hematopoietic stem cell transplantation.

Keywords: Bone marrow stromal cells; Cellular therapy; Clinical translation; Cryopreservation; Engraftment; Immunotherapy; Instant blood-mediated inflammatory reaction.

Figure 1

Freeze-thawed MSCs elicit an increased triggering of the instant blood mediated inflammatory reaction (IBMIR). Donor-matched freshly harvested or freeze-thawed MSCs (15,000 cells per milliliter) were tested for triggering of the IBMIR by exposing them to non-anticoagulated whole blood in the chandler blood loop model. (A): Representative photographs of clot formation after a 60-minute blood exposure of either freshly harvested or thawed MSCs, or buffer as negative control. (B): Percentage (relative to PBS, 30 minutes time point) of coagulation and complement activation markers after blood exposure with fresh or thawed MSCs (n = 30): free platelets and ELISA quantification of TAT, complement C3 activation fragment a (C3a), and soluble C5b-9 complex (sC5b-9). The dotted line corresponds to values obtained for PBS-treated blood. Boxplot whiskers 1.5 interquartile range, *, p<.05, **, p<.01; ***, p<.001, paired t test. Abbreviations: MSCs, mesenchymal stromal cells; PBS, phosphate buffered saline; TAT, thrombin-anti-thrombin complex.

Figure 2

Freeze-thawed MSCs exhibit increased sensitivity to complement lysis. Time lapse live cell imaging (A, B) and electrical-impedance based CASY cell counter (C, D) were used to study the cell morphology and viability of fresh or freeze-thawed MSCs after exposure to either PBS (no serum), complement active NHS, or EDTA-inactivated NHS/EDTA without complement lysing activity. (A): Transmitted light images (5, 30, and 60 minutes) for fresh or thawed MSCs within focal plane for one representative MSC donor after respective serum treatments. (B): Quantification of the average visible cell number within focal plane of microscope (n = 6 MSC donors, data are presented as percentage of fresh non-serum-treated cells at time point zero set to be 100%, corresponding videos can be found in supplement). (C): Cell viability of cells in suspension (% of counted cells, n = 9 MSC donors) and (D) total viable cell number in suspension (% of parent population at start, n = 9 MSC donors). Mean ± SD, *, p < .05; **, p <.01, p values from paired t test. Abbreviations: MSC, mesenchymal stromal cells; NHS, normal human serum.

Figure 3

Freeze-thawed MSCs show increased cell shrinkage, debris formation, propidium iodide (PI)-membrane permeabilization, and AV-binding after serum exposure. Flow cytometry study of cell morphology and viability of freshly harvested or freeze-thawed MSCs after exposure to either PBS (none), complement active NHS, or EDTA-inactivated NHS/EDTA. (A, B): Typical dot plots indicating cell size (FSC) and granularity (SSC) of freshly harvested and freeze-thawed MSCs within gate R0 before and after serum treatment, with indication of cell shrinkage (upper arrow) and formation of cell debris (lower arrow); quantitative results (n = 9 experiments) are shown in panel (B). (C, D): Histogram overlays for detection of PI incorporating cells (gate R1, PI-labelled cells shown as red histograms) within gated cell population (gate R0) versus unlabeled negative control (grey histogram), with indication of increased PI-incorporation (black arrows) in NHS-treated cells (n = 9 MSC donors, with 5 technical replicates in each experiment). Cell viability presented in panel (D) was calculated based on PI-exclusion. If cell count stopped before set threshold (2,000 counts) missing counts were counted as dead cells. (E, F): Histogram overlays for AV binding to MSCs (shown as green histograms) versus negative control (shown as grey histograms) and quantification of AV-positive cells (gate R2), with indication of increased AV-binding to NHS-treated cells (black arrows). MSCs were subjected to similar treatment as described above (n = 6, one sample each), and the percentage of AV-positive cells (apoptotic cells) is shown in panel (F). The data in (B, D, and F) are presented as the mean ± SD. *, p < .05; **, p < .01, p values from paired t test. Abbreviations: AV, Annexin V; FSC, forward scatter; MSC, mesenchymal stromal cell; NHS, normal human serum; SSC, sideward scatter.

Figure 4

Freeze-thawed MSCs have impaired immunomodulatory properties. (A): IDO mRNA expression (fold change) in freshly detached or freeze-thawed MSCs (n = 4 each) was studied with quantitative reverse transcriptase polymerase chain reaction (QRT-PCR) analysis after 24-hour stimulation with or without IFN-γ (100 U/ml), boxplot whiskers min to max, p value from Wilcoxon signed rank test. (B): Quantification of IDO enzymatic activity in MSC conditioned medium (n = 4). IDO activity was detected as tryptophan metabolite L-Kynurenine (μM). Means ± SD, p values from paired t test. (C, D): Suppression of peripheral blood mononuclear cell (PBMC) proliferation (% proliferation of PBMCs in presence of MSCs) by freeze-thawed or fresh MSCs (n = 12 tests in total, with n = 6 MSC donors, and n = 2 PBMCs) is shown relative to dividing control PBMCs without MSCs (100% proliferation). The suppressive effect of MSCs was studied either with phytohaemagglutinin (PHA)-mitogen- (C; readout day 3–4), or alloantigen-stimulated (D; readout day 5–6) mixed lymphocyte reactions. Boxplots whiskers min to max, *, p<.05; **, p<.01, paired t test. Abbreviations: IDO, indoleamine 2,3-dioxygenase; IFNg, interferon gamma; MSC, mesenchymal stromal cell; PBMC, peripheral blood mononuclear cell; PHA, phytohaemagglutinin.

Figure 5

Analysis of patient response to fresh and thawed MSCs. (A): Between 2002 and 2007, nine infusions of fresh MSCs were given to patients with treatment indication aGvHD and HC, which was paralleled by application of 35 freeze-thawed cells to patients with similar indication in the same time frame. The average response to all 44 infusions was 60%. Early passage cells (P1–2) yielded a 71% response when given thawed (n = 7) and a 100% response when given fresh (n = 5). (B): Evaluation of clinical response to MSCs for stratified patient cohort regarding cell viability at infusion (%). MSCs given to complete and partial responders (CR and PR, n = 13) showed similar viability (TRYPAN, median viability 93% vs. 90%) as cells given to patients with stable and progressive disease (SD and PD, n = 23). Boxplot whiskers min to max. Abbreviations: aGvHD, acute graft versus host disease; CR, complete responder; HC, hemorrhagic cystitis; MSC, mesenchymal stromal cell; PD, progressive disease; PR, partial responders; SD, stable disease.