Recent reports demonstrated that dECM deposited by synovium-derived stromal/stem cells (SDSCs) provided an in vitro microenvironment for SDSC expansion, which dramatically improved proliferation and enhanced chondrogenic potential ^{14– 16} . The flow cytometry data reported by Li et al. showed that, compared with surface markers of human SDSCs grown on tissue culture plastic (TCP), the percentage of CD29, CD90, and CD105 expression of SDSCs grown on dECM decreased slightly but the median fluorescence intensity (MFI) declined dramatically; interestingly, both the percentage and MFI of stage-specific embryonic antigen 4 (SSEA4) increased ¹⁷ . Zhang et al. also found that, despite nearly 100% expression in SDSCs after expansion on either TCP or dECM substrates, CD29, CD90, and CD105 declined dramatically at the MFI in dECM-expanded SDSCs; the MFI of SSEA4 in dECM-expanded cells increased slightly while the percentage doubled ¹⁸ . Interestingly, real-time quantitative polymerase chain reaction results showed that SRY-Box9 ( SOX9), type II collagen ( COL2A1), and aggrecan ( ACAN) were significantly upregulated during chondrogenic induction in SDSCs from the dECM group ¹⁸ . In those two studies, SSEA4 was found to be the only surface marker under evaluation that increased when dECM improved the chondrogenic potential of human SDSCs. However, Li et al. showed that SSEA4(+) expression did not favor human SDSC chondrogenesis because enhanced chondrogenesis occurred in the SSEA4(−) population of cells ¹⁹ .

In 2011, Kim et al. found that the expression percentage of surface marker CD49a in human SDSCs decreased with pretreatment using FGF2 and that FGF2 exerted no effect on the expression levels in CD29, CD44, CD73, CD105, and CD166 ²⁰ . In that study, the size, weight, and glycosaminoglycan (GAG) accumulation of pellets increased following FGF2 supplementation during cell expansion. In another study, after seven days of monolayer expansion in the presence of FGF2, human SDSCs became significantly smaller and showed a fibroblast-like appearance as well as a decrease in MFI for CD29, CD90, and CD10; however, FGF2-pretreated SDSCs showed significantly increased chondrogenic potential ²¹ . Hagmann et al. reported that FGF2 pretreatment suppressed CD146 expression in human bone marrow–derived stromal/stem cells (BMSCs) and promoted chondrogenic differentiation ²² . These studies demonstrated that, during ex vivo expansion, FGF2 is an effective agent to promote human MSC proliferation and chondrogenic potential via upregulation of SOX9 ²³ . However, measured surface markers did not show a positive correlation with the proliferative and chondrogenic potential of MSCs.

CD271, a low-affinity nerve growth factor receptor, is considered to be a highly selective surface marker for BMSCs ²⁴ . The reports from Mifune et al. ²⁵ and Calabrese et al. ²⁶ showed that CD271(+) BMSCs from freshly isolated cells had higher chondrogenic potential as evidenced by increased expression of chondrogenic genes in pellet culture with induction medium compared with CD271(−) BMSCs. Petters et al. demonstrated that, without ex vivo expansion, human bone marrow–derived CD271(+) mononuclear cells could generate sufficient articular cartilage constructs exhibiting high cell viability and remarkable chondrogenic matrix deposition in a type I collagen hydrogel ²⁷ .

Given that CD271 plays an important predictive role in chondrogenic potential of human BMSCs, there is still controversy over whether human SDSCs express CD271 ^{28, 29} . Some studies reported that CD271 was expressed only in the synovial membrane of patients with osteoarthritis ^{29, 30} . Interestingly, increasing evidence indicates that SDSCs are tissue-specific stromal/stem cells for chondrogenesis ³¹ and present superior chondrogenic potential and less hypertrophy compared with BMSCs ³² .

Despite the low abundance of CD271(+) subpopulation within stromal vascular fraction cells, Quirici et al. found that the CD271(+) subpopulation of adipose-derived stromal/stem cells (ADSCs) had high increments in cell proliferation when compared with unsorted ADSCs ³³ . Research by Kohli et al. showed that CD271(+) ADSCs from ex vivo expansion had a superior ability to promote cartilage repair compared with unsorted ADSCs ³⁴ . However, the study by Beckenkamp et al. showed that, in freshly isolated cells, CD34(+)CD271(+) ADSCs displayed similar in vitro chondrogenic potential at passage 3 compared with CD34(+)CD271(−) ADSCs ³⁵ .

CD105 (endoglin) is a transmembrane protein that regulates cellular proliferation, differentiation, and migration ³⁶ . Cleary et al. found that the percentage of CD105 in human BMSCs was not related to subsequent chondrogenic potential since CD105 expression did not change during cell expansion when chondrogenic potential decreased ³⁷ . Interestingly, this outcome seems to be inconsistent with the role of CD105 in SDSCs and ADSCs. Arufe et al. in 2009 ³⁸ and Chang et al. in 2013 ³⁹ demonstrated that the cellular subset of CD105(+) SDSCs from ex vivo expansion possessed greater chondrogenic capacity than the CD105(−) SDSC subset. Jiang et al. found that CD105(+) ADSC subpopulation in in vitro culture had a much stronger chondrogenic potential than CD105(−) subpopulation and had more intensive immunostaining of type II collagen and higher gene expression of COL2A1 and ACAN following chondrogenic induction ⁴⁰ . There is also a report that myrtucommulone-A treatment reduced CD105 expression in expanded human ADSCs along with reduced chondrogenic potential ⁴¹ .