Numerous clinical trials are currently being performed using AMSCs as a cellular therapy for various diseases worldwide [32, 33]. Standardization of the production procedures and accurate characterization of the MSC product to ensure patient safety has been a significant concern for regulatory agencies governing the approval of biological license applications [2]. Our study identified and validated the expression of 14 classical and non-classical surface markers on clinical-grade AMSCs expanded in hPL adherent to good manufacturing practices (GMP-hPL). Furthermore, we evaluated surface marker expression during processes for preparing cells for clinical administration and demonstrated variability of these markers with doublings/day and cryopreservation.

Traditionally AMSCs have been expanded with FBS as part of the culture media to provide growth factors and other proteins to support proliferation. However, potential zoonotic pathogens and immunogenic reactions from FBS are concerning for the clinical administration of MSCs, which led to the development of nonzoonotic substitutes including hPL [12, 34]. As part of the production of clinical-grade AMSCs used in these studies, we expanded our MSC product in GMP-hPL, which has previously been shown to support proliferation and genomic stability [12]. Furthermore, hPL contains proteins important for healing, including FGF/EGF, TGF-β/BMP, and VEGF/PDGF, which may facilitate AMSC growth and stability [35]. However, due to the differences in composition of FBS and hPL, including cytokines and growth factors, there exists the potential for the selection of different adherent cell populations.

An AMSC cell population is characterized as one that adheres to plastic, expresses characteristic surface markers, and has tri-lineage potential [11]. Establishing these criteria was an important step forward for standardization of stem cell science and industry. This study validated that clinical-grade AMSCs from 15 different donors met these established criteria and also expressed a unique set of non-classical surface markers. The AMSCs utilized in this investigation were cultured in GMP-hPL, adhered to plastic, and uniformly expressed the classical surface markers CD44, CD73, CD90, and CD105 and did not express CD34. Although these markers uniformly define AMSCs, ours and other studies have observed that these markers are unable to distinguish donor differences, including variability in proliferation or trophic activity [36]. Therefore, there is an increased need to identify additional markers that not only define AMSCs but also have the potential to capture biological and manufacturing variability, as well as clinical performance.

Beyond serving as markers for cell characterization, surface proteins carry out important biological functions and are critical for cell-to-cell contact, extracellular matrix interactions, signal transduction, and transportation of molecules across the plasma membrane. Our study examined the expression of all plasma membrane protein-encoding genes (including CD markers, receptors, integrins, and transporters) by RNA-seq and identified 551/707 genes that were expressed on AMSCs. Previous studies of AMSCs and BMSCs evaluated the expression of 200–242 surface markers using mass spectroscopy or BD Lyoplate technology with flow cytometry techniques [2124]. Together, our and others’ studies have identified and characterized a finite number of surface proteins present on the AMSC cell surface under standard conditions. A comparison of surface marker expression by qPCR expression and flow cytometry showed partial concordance, with seven out of nine markers showing similar trends (Fig. 4). Our results show that gene expression and flow cytometry techniques can be used to identify novel cell surface markers in AMSC populations. However, confirmation of cell surface proteins by flow cytometry is still necessary to confirm cell surface marker expression, as it is the primary technique used in the clinical setting.

Our data also reveal differential expression of mRNAs for CD markers in proliferative and post-proliferative AMSCs. Previous studies have also characterized the gene expression effects of confluence and doubling times on AMSCs and BMSCs [13, 37]. These studies and ours suggest that culture conditions, including proliferation state and population doublings, may affect the differentiation potential of AMSCs. In particular, mRNA expression of some surface markers is restricted to either the proliferating or confluent state (Fig. 6c). However, further studies are required to determine whether manufacturing conditions (e.g., length of culture and growth rate), as well as biological factors (e.g., donor age and disease status) would impact therapeutic potential. Our studies indicate that at least some variation in cell surface expression may emerge during AMSC production depending on the growth rate of the cell population.

Currently there is no standardization of release criteria for MSC products, and studies that define their criteria usually include the markers described by the ISCT and IFATS [2, 38]. Through the use of gene expression profiling techniques and flow cytometry, our study identified and validated nine non-classical markers that may help further characterize hPL-expanded AMSCs and improve current release criteria. In our study we validated the expression of CD163, CD271, CD200, CD36, CD274, CD146, CD248, CD140B, and CD276. These markers are also expressed on FBS-expanded AMSCs [2225, 3942]. Our study also identified CD163, a monocyte and macrophage marker, as a negative AMSC marker, which may be useful for characterizing clinical-grade MSC populations [43].

We also described the expression of the known immune-regulatory markers CD274 (B7H1/PD-L1) and CD276 (B7H3) on hPL-expanded AMSCs. Traditionally, mature dendritic cells produce soluble CD274 and CD276 [44]. However, the current results demonstrate that AMSCs grown in hPL also express these markers on the cell surface. Our results show that CD276 is highly expressed and may be expressed ubiquitously with other traditional markers such as CD73, CD105, and CD90. Similarly, CD274 is highly expressed, but shows greater variability between donors. The functional role of CD274 and CD276 on AMSCs has yet to be characterized; however, CD274-positive BMSCs have been shown to regulate T-cell proliferation and Th17 polarization [45, 46]. Furthermore, recent studies have shown that interferon gamma (IFNγ) priming or licensing of BMSCs may also up-regulate CD274 and enhance MSC-mediated T-cell inhibition [47, 48]. However, the function of CD276 remains controversial as this molecule may act as a co-stimulatory molecule for T-cell activation and selectively stimulates the production of IFNγ [49], or may inhibit T-cell proliferation [50]. CD274 and CD276 have the potential to serve as predictive clinical markers for MSC immunomodulatory activity. Isolation and characterization of CD274-positive and CD276-positive cells may determine whether these cells represent distinct subpopulations with enhanced immune-regulatory effects. Further studies correlating these markers with patient outcomes in clinical trials would also help to elucidate the role of these markers in hPL-expanded AMSCs.

Storage and administration of the MSC product for cell therapy may depend on the disease and the institutional infrastructure. Current clinical trials administer MSC products either without cryopreservation, cryopreserved and thawed, or allowed to recover for 4 days in culture. Previous studies have demonstrated reduced immunosuppressive properties of MSCs immediately thawed after cryopreservation, and that these properties were restored as early as 24 h after placing in tissue culture [51]. Samples from three of our five donors analyzed showed a slight decrease in CD248 expression between pre-freeze and post-thaw samples. The decrease in surface marker expression could be attributed to damage to the cell surface of the protein that reduces antibody-binding efficiency or, potentially, the sensitivity of CD248 expression to the metabolic state of the cell. We also observed a significant increase in CD105 expression between pre-freeze and post-thaw samples, as well as a significant increase in surface marker expression over 4 days for CD271, CD36, and, to a lesser extent, CD44. These results support the work of Francois and colleagues [51], whereby the recovery of AMSCs in culture for up to 4 days can result in maximal surface marker expression. Together, these data suggest that surface marker expression is modulated during the cryopreservation process and that it may be important for cell function to allow cells to recover for up to 4 days.