TABLE 2.

Comparison of different studies reporting static and dynamic methods of spheroid formation and their impact on the CM profile.

References	MSCs source	Formation method	Viability and proliferation	CM characterization and secretion of paracrine factors	Other results
Frith et al. (2010)	human BM-MSCs	Static (ex situ) Microwells in non-adherent plates	Most of the cells were viable, with no differences between cells on the periphery and those in the centre; 3.9, 4.3 and 3.5% of cycling cells for monolayer, SF and RWV culture.	CM obtained from MSCs cultured in monolayer or stirring flasks (3D) was used to treat both MSCs and other cells from the bone marrow microenvironment and viability was determined by MTT assay. The results showed no difference in the viability of primary MSCs, C3H101T1/2 or human umbilical vein endothelial cells cultured in 2D and 3D CM. However, the viability of the human osteosarcoma cell line Saos-2 was significantly decreased in cells from 3D versus 2D CM. No secretion of paracrine factors reported.	Increased osteogenic and adipogenic differentiation potential of MSCs spheroids compared with MSCs cultured at 2D.
Bhang et al. (2011)	human AD-MSCs	Dynamic (in situ) Single-cell inoculation in SF	Cell adhesion and migration were preserved.	Increased secretion of angiogenic factors (VEGF, HGF, FGF2 and CXCL12) from hADSC cultured as spheroids versus hADSC cultured at monolayer. HGF: 750 vs 200 [pg per 104 cells], VEGF: 780 vs 400 [pg per 104 cells] and FGF2: 500 vs 250 [pg per 104 cells].	Transplantation of spheroids promoted angiogenesis in mouse ischaemic limb tissue.
Hildebrandt et al. (2011)	human BM-MSCs	Dynamic (in situ) Non-adherent bacterial plate on a rotating platform	Very low viability estimated by PI/FDA staining	No secretion of paracrine factors reported.	Very heterogeneous spheroids in size and number. Aggregates form larger amorphous masses, with low viability after 1 week.
Hildebrandt et al. (2011)	human BM-MSCs	Static (ex situ) HP	Small aggregates are viable, estimated by PI/FDA staining	No secretion of paracrine factors reported.	Increased multiaggregation compared with spheroids formed by HP.
Hildebrandt et al. (2011)	human BM-MSCs	Dynamic (in situ) Non-adherent bacterial plate on a rotating platform	High viability estimated by PI/FDA staining. However, a clear tendency towards decreased viability at day one by WST-1 assay.	No secretion of paracrine factors reported.	Increased efficiency of formation and more controlled size compared with spheroids cultured under dynamic conditions, at lower initial cell density.
Baraniak & McDevitt, (2012)	Murine BM-MSCs	Static (ex situ) Microwells non-adherent plates	Absence of necrotic core. BrdU staining confirmed the retention of MSC proliferative capacity after dynamic culture.	No secretion of paracrine factors reported.	Increased adipogenic and osteogenic potential of cells recovered from 3D cultures. Maintenance of MSCs plasticity following 3D culture.
Bhang et al. (2012)	human CB-MSCs	Static (ex situ) HP	MTT assay showed that cell viability is higher in spheroids than in 2D cultures. No proliferation was reported.	Increased secretion of HGF, VEGF and FGF2 from hADSC cultured as spheroids versus hADSC cultured at monolayer. Values not reported.	Increased expression of Bcl-2 in spheroids cultured under hypoxia compared with that in 2D cultures under normoxia and hypoxia or 3D cultures under normoxia
Cho et al. (2012)	human AD-MSCs	Static (ex situ) HP	nr.	Increased secretion of TGFβ1 and VEGF of CM obtained from 3D cultures compared with baseline culture medium (αMEM). VEGF: 1,015.17 ± 170.97 [pg/ml], TGFβ1: 14.33 ± 6.71 [pg/ml]. Increased in vitro pro-angiogenic effects of CM obtained from 3 days cultures confirmed by a tube-formation assay. The continuous infusion of CM obtained from 3D cultures induced significantly better functional and structural recovery after stroke, reducing the infarction volume and maintained motor function in an ischemic stroke model.	-
Alimperti et al. (2014)	human BM-MSCs	Dynamic (in situ) Single cell inoculation in SF	Spheroids showed a 80% of viability until day 5. Cells were quantified with an automatic cell counter Vi-CELL, which measures viable cell density, viability and average cell size.	No secretion of paracrine factors reported.	High levels (>99%) of MSC surface markers. Increased trilineage differentiation potential for spheroids cultured in serum-free medium.
Bhang et al. (2014)	human AD-MSCs	Static (ex situ) HP	3D spheroid culture system was able to support the growth of cells at a density app. 4 times higher than that observed for 2D cultures, for both media.	CM obtained from spheroids culture had a significantly higher concentration of angiogenic factors than the monolayer culture CM (VEGF, FGF2, HGF, and CXCL12 in the αMEM (serum+) spheroid culture CM was 14.4 ± 0.4, 13.2 ± 2.2, 13.3 ± 2.3, and 16.6 ± 2.9 [ng/ml], respectively). In vitro and in vivo antiapoptotic effect, was observed in an ischemic hindlimbs model. In addition, in vivo angiogenic effect of CRM-bases spheroid CM. and an improved blood perfusion in the ischemic limbs.	Spheroids cultured in CRM (without serum) supported culture at a significantly higher maximal cell density compared with the monolayer culture supplemented with serum (×105 cells/ml; 7.0 ± 0.8 versus 3.1 ± 0.5). Serum deprivation caused CASP3 pathway activation and increased TP53 mRNA expression, regardless of the type of medium or culture system used.
Zimmermann and Mcdevitt, (2014)	human BM-MSCs	Static (ex situ) Microwells in non-adherent plates	$\uparrow$ no. of spheroids in Mesencult-XF In FBS medium, no change in the total number of cells in spheroid cultures occurred after 4 days, indicating no significant expansion in MSCs over the culture period. A 1.9, 2.0 and 2.9-fold change in the total number of cells was observed in 200-cell, 500-cell and 1000-cell spheroid culture in MesenCult-XF medium after 4 days.	Increased secretion of PGE2, TGFβ1, and IL6 from spheroids hMSC, compared with human MSCs cultured. Increased secretion of IL6 in spheroids cultured in MeseCult-XF medium compared with cells grown in 2D culture, which did not secrete detectable levels of IL6. Increased secretion of immunomodulatory factors by 500-cell spheroids	-
Kwon et al. (2015)	human AD-MSCs	Dynamic (in situ) Single-cell inoculation in SF	nr.	Monolayer cultured hADSCs in αMEM medium without supplemental serum or supplements, secreted VEGF (0.56 ± 0.22 ng/ml), FGF2 (0.51 ± 0.06 ng/ml), HGF (0.55 ± 0.08 ng/ml), and CXCL12 (0.085 ± 0.07 ng/ml). When hADSCs were cultured in spheroid culture, significant increases in VEGF (12.3 ± 2.4 ng/ml), FGF2 (11.0 ± 1.7 ng/ml), HGF (10.8 ± 3.6 ng/ml), and SDF-1a (12.5 ± 3.8 ng/ml) were observed. The concentration of growth factor per cell in spheroid culture CM was approximately 20-fold higher for VEGF, FGF2, and HGF and 145-fold higher for CXCL12 as compared with the monolayer culture CM.	Increased cell density in 3D (10.6 × 105 cells/ml) vs. 2D culture (2.95 × 105 cells/ml).
Li et al. (2015)	human UC-MSCs	Dynamic (in situ) Single-cell inoculation in dishes on a rocker system	Over 95% at day 9. Ki-67 staining showed that the cells retained their ability to proliferate.	No secretion of paracrine factors reported.	Increased expression of Oct4, Nanog, Sox2 and Rex1, with 4.3, 3.9, 6.2 and 3.2-fold increases vs. 2D culture.
Santos et al. (2015)	human UC-MSCs	Dynamic (in situ) Single-cell inoculation in SF	Absence of necrotic core at day 11. Ki67 staining showed the presence of proliferating cells. However, Ki67 positive cells comprised only a small fraction of cells, indicating that only a low fraction (<5%) of cells were actively proliferating in spheroids.	Increased secretion of HGF, TGFβ1, FGF2, IL6, and GCSF in CM obtained from 3D cultures than in CM from 2D cultures. Most impressively, VEGFA, which was only residually secreted in 2D cultures, was highly secreted by MSCs under 3D conditions (80-fold higher than CM obtained from 2D). The results strongly suggested an improved paracrine effect of CM obtained from 3D cultures onto fibroblast-mediated ECM synthesis, angiogenesis and vasculogenesis, essential for the granulation tissue formation and remodelling stages of wound healing.	From day 6 onwards, the population of 3D spheroid-dissociated cells showed a decrease in CD105 and C90 expression levels that restored once spheroids were plated back. MSCs grown in 3D cultures were app. 30% smaller in size when compared to cells grown in 2D. In addition, MSCs retained the ability to adhere and proliferate on plastic surface and tridifferentiation potential.
Zhang et al. (2015)	human AD-MSCs	Dynamic (in situ) RWV	Most of the cells were viable. Absence of necrotic centre. Spheroid-derived ADSCs exhibited significantly stronger proliferative ability than cells grown in 2D culture at later time points.	In addition, the concentration of growth factors per cell in spheroid culture CM was 20-fold higher than that in 2D culture CM for VEGF, FGF2, and HGF and 145-fold higher for CXCL12.	Increased expression levels of Oct4, Nanog, Sox2, and Rex1 compared with those in 2D culture.
Costa et al. (2017)	human BM-MSCs	Static (ex situ) Microwells in non-adherent plates	Similar cell viability values were found for both monolayers and spheroids.	Increased secretion of HGF and VEGF. .CM obtained from 3D cultures exhibited higher closure of the wounded area 8 h after (app. 40%) relatively to monolayer-derived CM (app. 27%) in an in vitro scratch wound healing assay.	Spheroids exhibited increased resistance to oxidative stress compared t single MSCs. Increased expression level of TSG6.
Egger et al. (2017)	human AD-MSCs	Dynamic (in situ) Single-cell inoculation on STR	78.5% (normoxic) and 86% (hypoxic) viability; 1.85 - fold (normoxic) and 2.23 -fold (hypoxic) cell expansion.	No secretion of paracrine factors reported.	Surface markers of cells cultivated under normoxic and hypoxic conditions were comparable and met the minimal criteria of MSCs.Increased adipogenic and chondrogenic differentiation under 3D hypoxic culture vs. 3D normoxic culture.Decreased osteogenic differentiation under 3D hypoxic culture vs. 3D normoxic culture.Glucose consumption (0.85 ± 0.1 mmol) and lactate production (1.69 ± 0.11 mmol) were significantly lower in normoxic conditions compared to hypoxic conditions, where glucose consumption was 1.09 ± 0.02 mmol and lactate production 2.05 ± 0.09 mmol.
Cha et al. (2018)	human BM-MSCs	Static (ex situ) Microwells in non-adherent plates	Live/Dead assay showed that most cells in the spheroids were viable during the culture period. No changes in spheroid numbers.	Increased production of MV. Highest enrichment of hMSC-derived MVs was found in dynamic 3D cultures, which was approximately 100-fold higher than in the 2D control containing only a few secreted MVs.	Upon formation of hMSC-spheroids at day 1, GDF15 and TGFB3 were upregulated by approximately 40-fold compared to the 2D control, whereas BMP4 was downregulated by approximately 60-fold. At day 7, IL1B, BDNF, and BMP2 were upregulated by over 30-fold while COL1A1 was downregulated by approximately 50-fold, compared to the early stage on day 1.
He et al. (2019)	rabbit BM-MSCs	Dynamic (in situ) Single-cell inoculation on SF	nr.	No secretion of paracrine factors reported.	After 24 h of culture, approximately 80% of rMSCs were incorporated into cellular aggregates. Faster aggregation at a lower agitation rate and a higher cell inoculation density.
Allen et al. (2019)	human SyF-MSCs	Static (ex situ) Microwells in non-adherent plates	Proliferation ceased at day 6.	No secretion of paracrine factors reported.	Increased collagen production in spheroids. Highly variable size of spheroids in dynamic culture.
Allen et al. (2019)	human SyF-MSCs	Dynamic (in situ) Single-cell inoculation in a STR	Proliferation ceased at day 6.	No secretion of paracrine factors reported.	Single-cell inoculation yields a more uniform population of smaller aggregates after eight days of culture. Single-cell and preformed spheroid inoculation achieved similar fold changes in cell numbers and SGAG.
Miranda et al. (2019)	human UC-MSCs	Dynamic (in situ) Single-cell inoculation in SF	Spheroids were viable until day 7 given haematoxylin and eosin staining images.	Increased secretion of anti-inflammatory cytokines such as IL10 and LIF, as well trophic factors involved in different mechanisms leading to tissue regeneration, mainly PDGFB, FGF2. CCL1 SCF and GMCSF in CM obtained from 3D cultures. CM derived from 3D dynamic cultures promoted a 1.5-fold increase in chondrocyte migration capacity 24 h post-scratch in an in vitro healing assay, when compared to CM obtained from 2D cultures Also, results showed that CM obtained from 3D cultures has a clearly superior capacity for both, avoiding and ameliorating adjuvant induced arthritis (AIA) manifestations in vivo when compared to CM obtained from 2D cultures or even MSCs. CM treatment was able to both prevent a reverent all major signs of AIA, including complete avoidance of necrotic foci around joints, acute and chronic inflammation, joint deformity and secondary infection.	-
Niibe et al. (2020)	human BM-MSCs	Dynamic (in situ) Single-cell inoculation in SF	Live/dead staining after 4 weeks dynamic 3D culture showed that most cells on the spheroids surface and outer layer were alive.	No secretion of paracrine factors reported.	Decreased expression level of Ki67, known as a market of cell proliferation, suggest that the dynamic 3D culture mimicked a state of quiescence or stopped the MSCs cell cycle.
Niibe et al. (2020)	human BM-MSCs	Dynamic (in situ) Single-cell inoculation in SF	Live/dead staining after 4 weeks dynamic 3D culture showed that most cells on the spheroids surface and outer layer were alive.	No secretion of paracrine factors reported.	Increased expression level of CD27 in spheroids.