Esta referencia está disponible actualmente en inglés. La versión en español está en preparación.

MSC Licensing and Preconditioning: IFN-γ, TNF-α and Hypoxia

Por Benn Bluestein-Veyra, M.Sc. en Química Orgánica · Actualizado el 20 de octubre de 2026

MSC licensing is the exposure of mesenchymal stromal cells to a defined signal that switches on an inducible gene program before the cells are assayed. The best-studied signals are IFN-gamma and TNF-alpha, low oxygen culture and aggregation into three-dimensional spheroids.^[1][2][3]

This reference covers each priming condition, the genes and secreted factors it induces, and the assays that measure the response. The molecular actions of IDO1, PGE2 and TSG-6 belong to the MSC immunobiology reference, and this page centers on the culture conditions that set them in motion.

Defining MSC licensing as a response to the culture environment

Krampera and colleagues showed in 2006 that bone marrow MSCs suppressed the proliferation of CD4+ and CD8+ T cells and NK cells, and that this suppression required IFN-gamma produced by the activated lymphocytes. IFN-gamma acted by inducing IDO activity in the stromal cells.^[1]

Ren and colleagues reached a parallel result in mouse MSCs. IFN-gamma together with TNF-alpha, IL-1alpha or IL-1beta induced chemokines and inducible nitric oxide synthase, and the suppressive effect required the IFN-gamma receptor IFNGR1.^[4]

In the Krampera study, suppression worked without cell contact, and added IFN-gamma made even activated B cells susceptible to it.^[1] In the Ren study, the induced chemokines drew T cells close to the stromal cells, where nitric oxide held their responsiveness down.^[4]

These studies framed MSC immune activity as an inducible state set by the culture environment (Fig. 1). An ISCT working proposal led by Krampera then called for standardized immune assays that yield shared, reproducible and consistent data on that state.^[5]

Licensing, priming and preconditioning describe one practice

Three overlapping terms describe this practice. Licensing and priming usually refer to cytokine exposure, and preconditioning often refers to a period of hypoxia before a test; Cheng and colleagues group cytokines, hypoxia and chemicals together as priming approaches.^[2][6][7]

IFN-gamma and TNF-alpha priming induces a defined set of genes

IFN-gamma signals as a homodimer that engages two chains of its high-affinity receptor. Walter and colleagues solved this complex in 1995 and found one IFN-gamma homodimer bound to two receptor molecules held 27 Å apart.^[8] The human genes IFNG and IFNGR1 lie on chromosomes 12q15 and 6q23.3.^[9][10]

Thiel and colleagues solved the human complex with glycosylated receptor at 2.9 Å resolution (Fig. 2). Beside the expected 2:1 complex, the crystal held a third receptor molecule packed against the edge strands of the receptor C-terminal domains.^[11]

Kim and colleagues traced IDO expression in IFN-gamma-primed human MSCs to the JAK and STAT1 pathway. In their account, IFN-gamma from activated T cells induces IDO in the stromal cells, which in turn suppresses T cell proliferation.^[12]

Crop and colleagues profiled the full genome of adipose MSCs exposed to IFN-gamma, TNF-alpha and IL-6 or to a mixed lymphocyte reaction. The cytokines mainly raised IDO, the mixed lymphocyte reaction raised COX-2, and both conditions enhanced suppression with differentiation capacity preserved.^[13]

Roemeling-van Rhijn and colleagues found that IFN-gamma with TNF-alpha strongly raised IDO and PD-L1 in adipose MSCs. L-kynurenine accumulated in the medium as a direct measure of IDO activity, and the same induction appeared at 1% and 20% oxygen.^[14]

Hypoxic preconditioning stabilizes HIF-1alpha

Wang, Semenza and colleagues identified HIF-1 in 1995 as a heterodimer of two basic helix-loop-helix PAS proteins, HIF-1alpha and HIF-1beta, the latter encoded by ARNT. Both subunits rose in cells held at 1% oxygen and decayed rapidly on return to 20%.^[15]

Jaakkola and colleagues showed that oxygen controls HIF-1alpha through hydroxylation of proline 564, which lets the pVHL E3 ligase complex bind and ubiquitylate it (Fig. 3). The hydroxylase needs dioxygen and iron, so it acts as a direct cellular oxygen sensor.^[16]

Kaelin and Ratcliffe describe the PHD family of prolyl hydroxylases as the oxygen sensors of this pathway, with asparagine hydroxylation by FIH1 tuning HIF transcriptional activity further.^[17] The human HIF1A gene lies on chromosome 14q23.2.^[18]

Crystal structures show ARNT wrapped around HIF-1alpha

Wu and colleagues solved mouse HIF-1alpha and HIF-2alpha in complex with ARNT, including HIF-1alpha and ARNT on hypoxia response element DNA (Fig. 4). ARNT spirals around the outside of each HIF-alpha subunit, and the DNA-reading head cooperates with a distal PAS domain to bind the element.^[19]

HIF-1alpha and HIF-2alpha share this highly integrated quaternary architecture with ARNT. The structures also revealed five pockets able to bind small molecules, including sites inside the PAS domains and a cavity formed at the heterodimer interface.^[19]

Low oxygen culture changes growth, survival signaling and immune function

Roemeling-van Rhijn and colleagues note that oxygen levels within tissues are typically much lower than those of standard culture.^[14] Hypoxic preconditioning sets the culture oxygen closer to that tissue range before the cells are tested.^[2]

Grayson and colleagues kept human MSCs at 2% oxygen for up to seven passages and measured about 30-fold higher expansion over 6 weeks, with multilineage differentiation retained. Hypoxic cells also expressed more Oct-4, HIF-2alpha and connexin-43 and built more extracellular matrix.^[20]

Tsai and colleagues found that low-density hypoxic culture preserved MSC properties through a HIF-1alpha and TWIST axis that lowered E2A and p21. Hypoxic cells reached up to 100 population doublings with telomerase activity, stable telomere length and a normal karyotype.^[21]

Bader and colleagues preconditioned umbilical cord blood MSCs in hypoxia for 24 h with immunophenotype and karyotype unchanged. Preconditioned cells showed more AKT and BAD phosphorylation, higher BCL-XL, BAG1 and VEGF expression and more VEGF secretion, and they raised endothelial cell proliferation and migration.^[2]

Hypoxia preserves the cytokine licensing response

At 1% oxygen, adipose MSCs kept their proliferation and colony-forming efficiency, with CD105 the one marker that fell. IDO and PD-L1 induction by IFN-gamma and TNF-alpha, kynurenine output and suppression of CD4+ and CD8+ T cell proliferation all held steady.^[14]

Three-dimensional spheroid culture self-activates stromal cells

Bartosh and colleagues aggregated human MSCs into spheroids by hanging drop culture and found high expression of TSG-6 and stanniocalcin-1, with properties that depended critically on the culture conditions. Spheroid cells were about one-fourth the volume of cells from adherent monolayers.^[22]

In co-culture with LPS-activated macrophages, spheroid MSCs suppressed inflammatory responses more effectively than monolayer MSCs.^[22] Ylöstalo and colleagues traced the activity of spheroid conditioned medium to PGE2, which required COX-2 and acted on macrophages through the EP4 receptor.^[23]

Spheroid medium lowered TNF-alpha, IL-6 and IL-12p40 secretion by stimulated macrophages and raised IL-10, IL-1Ra and CD206. PGE2 production depended on caspase activity and NF-kappaB activation, and fibroblast spheroids produced little PGE2.^[23]

Spheroid format and cytokine priming interact

Redondo-Castro and colleagues formed spheroids from 20,000 human MSCs and followed their compaction over 6 d (Fig. 5). Spheroid culture raised IL-1Ra, VEGF and G-CSF secretion compared with monolayers, and IL-1 priming acted more strongly on monolayers than on spheroids.^[24]

In the same study, spheroid conditioned medium lowered TNF-alpha secretion by LPS-stimulated BV2 microglial cells, and whole spheroids in co-culture raised IL-6. Protein arrays showed a stronger immune profile after priming, which the spheroid format partly offset through higher IL-6.^[24]

Cesarz and Tamama review the wider changes in spheroid MSCs: cytoskeletal reorganization, higher expression of pluripotency marker genes, delayed replicative senescence and broad shifts in gene expression measured by microarray. The authors identify the signaling pathways that trigger and sustain these changes as an open question for the field.^[3]

Priming reshapes the secretome and extracellular vesicle cargo

Harting and colleagues compared extracellular vesicles from naive and inflammation-stimulated MSCs, isolated with a tangential flow filtration system compliant with current Good Manufacturing Practice. The two preparations matched in size and common surface markers and differed in protein composition, cytokine profile and RNA content.^[25]

Vesicles from stimulated cells lowered pro-inflammatory cytokine release in vitro further than vesicles from naive cells. Activated leukocyte subsets took them up in a distinct pattern, and part of their activity traced to the COX-2 and PGE2 pathway.^[25]

Cheng and colleagues primed adipose MSCs with IFN-gamma and TNF-alpha and found higher RAB27B expression and more small vesicles under 200 nm. These vesicles carried A20 and TSG-6 and inhibited T cell proliferation in a dose-dependent manner.^[7]

In the same study, MSCs from pediatric donors expressed more RAB27B and secreted more small vesicles than adult MSCs.^[7] Peltzer and colleagues primed bone marrow MSCs from five donors with IFN-gamma or hypoxia and found vesicle amount and size unchanged.^[26]

Of 81 miRNAs detected in those vesicles, 41 appeared in every sample. Hypoxia lowered miR-34a-3p and IFN-gamma raised five miRNAs, including miR-126-3p and miR-451a, and the authors described the overall change as limited, with high variability between donors.^[26]

Gene panels and secretome assays measure the licensing response

Galipeau and colleagues, reporting an ISCT workshop, proposed that MSC potency be measured by licensing the cells in vitro and reading selected markers with a matrix of assays. The preferred methods were quantitative RNA analysis, flow cytometry of functional surface markers and protein assays of the secretome.^[6]

Chinnadurai and colleagues tested two such systems on MSCs from human donors: secretome analysis and a quantitative RNA array of immunomodulatory and homing genes. A cytokine signature in the secretome correlated with T cell suppression, and IFN-gamma served as a surrogate for responder blood mononuclear cells.^[27]

Single studies often read one licensed culture several ways. Roemeling-van Rhijn and colleagues measured IDO and PD-L1 expression, L-kynurenine in the medium and suppression of T cells stimulated with anti-CD3 and anti-CD28 antibodies.^[14]

Crop and colleagues paired whole-genome expression with functional assays of morphology, proliferation, differentiation and immune suppression in the same adipose MSCs.^[13] Kim and colleagues confirmed IDO’s role with knockdown experiments.^[12]

Where to go from here

The MSC immunobiology reference follows IDO1, PGE2 and TSG-6 from licensing to their molecular actions, and the secretome reference covers the soluble factors and vesicles that MSCs release into conditioned medium. The single-cell heterogeneity reference explains why donors and subpopulations answer priming differently.

The source tissues and ISCT minimal criteria references describe the cells these protocols start from, and the exosome biochemistry pages cover vesicle composition in depth. Verified practices can request lot documentation, and our team is glad to discuss the methods described on this page.