Human First-Trimester Fetal MSC Express Pluripotency Markers and Grow Faster and Have Longer Telomeres Than Adult MSC


  • Pascale V. Guillot Ph.D.,

    Corresponding author
    1. Experimental Fetal Medicine Group, Institute of Reproductive and Developmental Biology, Imperial College London, Hammersmith Campus, London, United Kingdom
    • Experimental Fetal Medicine Group, Institute of Reproductive and Developmental Biology, Imperial College London, Hammersmith Campus, Du Cane Road, London W12 0NN, U.K. Telephone: 44 (0)207-594-2121; Fax: 44 (0)207-594-2154
    Search for more papers by this author
  • Cecilia Gotherstrom,

    1. Experimental Fetal Medicine Group, Institute of Reproductive and Developmental Biology, Imperial College London, Hammersmith Campus, London, United Kingdom
    Search for more papers by this author
  • Jerry Chan,

    1. Experimental Fetal Medicine Group, Institute of Reproductive and Developmental Biology, Imperial College London, Hammersmith Campus, London, United Kingdom
    Search for more papers by this author
  • Hiroshi Kurata,

    1. Experimental Fetal Medicine Group, Institute of Reproductive and Developmental Biology, Imperial College London, Hammersmith Campus, London, United Kingdom
    Search for more papers by this author
  • Nicholas M. Fisk

    1. Experimental Fetal Medicine Group, Institute of Reproductive and Developmental Biology, Imperial College London, Hammersmith Campus, London, United Kingdom
    2. Centre for Fetal Care, Queen Charlotte's & Chelsea Hospital, London, United Kingdom
    Search for more papers by this author


The biological properties of stem cells are key to the success of cell therapy, for which MSC are promising candidates. Although most therapeutic applications to date have used adult bone marrow MSC, increasing evidence suggests that MSC from neonatal and mid-gestational fetal tissues are more plastic and grow faster. Fetal stem cells have been isolated earlier in development, from first-trimester blood and hemopoietic organs, raising the question of whether they are biologically closer to embryonic stem cells and thus have advantages over adult bone marrow MSC. In this study, we show that human first-trimester fetal blood, liver, and bone marrow MSC but not adult MSC express the pluripotency stem cell markers Oct-4, Nanog, Rex-1, SSEA-3, SSEA-4, Tra-1-60, and Tra-1-81. In addition, fetal MSC, irrespective of source, had longer telomeres (p < .001), had greater telomerase activity (p < .01), and expressed more human telomerase reverse transcriptase (p < .01). Fetal MSC were also more readily expandable and senesced later in culture than their adult counterparts (p < .01). Compared with adult MSC, first-trimester fetal tissues constitute a source of MSC with characteristics that appear advantageous for cell therapy.


Ideally, cell therapy strategies require stem cells endowed with pluripotential differentiation and self-renewal capacities that neither are tumorogenic nor raise ethical concerns associated with human ESC research. Human MSC are under investigation for applications in tissue engineering and regenerative medicine because they are relatively easy to isolate and can differentiate down mesoderm-derived lineages. Most studies to date have used adult bone marrow-derived MSC, although they are numerically rare, their numbers decrease with age, and they are slow to expand ex vivo. [1, [2]–3] MSC with similar cell surface markers and differentiation capacity have also been found in a range of developmentally younger tissues, such as the placenta [4], umbilical cord blood, and mesenchyme [5] and during mid-gestation from a variety of fetal tissues such as spleen, lung, pancreas, kidneys [6, [7]–8], and amniotic fluid [4, 9]. An accumulating body of literature suggests that these perinatal and mid-trimester MSC may have advantageous growth and plasticity properties over adult MSC, as they do in primates [10, 11]. Fetal MSC with potential for both autologous and allogeneic cell therapy have also been found earlier in gestation, circulating in human first-trimester fetal blood, and present in first-trimester liver and bone marrow [12]. All three fetal sources of first-trimester MSC have similar growth kinetics [12], whereas microarray data suggested that human fetal MSC may have a different growth potential compared with adult MSC [13].

Pluripotency indicates the capacity to differentiate into cell types of all the three germ layers. Study of its relationship with gene expression led to identification of several markers of pluripotent stem cells, such as the transcription factor Oct-4 [14, 15], which prevents stem cells from differentiating and is expressed in ESC, germ cells, and early whole embryos [16, 17]. Despite the fact that adult MSC possess broader differentiation capacity than originally thought, no consistent data indicate that they express pluripotency markers. Contrary to the expectation that only stem cells from embryonic tissues express Oct-4, however, it has recently been suggested that Oct-4 may be expressed in ontologically immature postembryonic stem cells, such as in second-trimester amniotic fluid [15, 18]. In terms of other pluripotency markers, Nanog is a homeodomain protein present in pluripotent human cell lines and absent from differentiated cells, which directs pluripotency and differentiation of undifferentiated ESC [19]. Pluripotent human stem cells can also be characterized by expression of SSEA-3 and SSEA-4, the keratan sulfate-associated antigens Tra-1-60 and Tra-1-81 [20], and Rex-1, all expressed at high levels in undifferentiated ESC and absent in differentiated cells [21].

Other factors relevant to cell therapy are self-renewal and senescence. Replicative stability is conferred by telomeres, double-stranded DNA (TTAGGG)n repeat sequences up to 20 kb long, with a single strand of the same sequence acting as a protective cap for the chromosomal ends. Because DNA polymerase cannot fully duplicate these sequences, telomeres shorten with successive cell division and DNA replication to reach a critical length, which triggers cell division arrest [22]. This replicative senescence [23] occurs after a finite number of cell divisions known as the Hayflick limit, which depends on the telomere length of the starting population [24]. ESC escape telomere shortening, as after each cell division, telomerase enzyme extends telomeres to compensate for sequence loss during replication [25, [26]–27]. Telomerase activity is closely related to the expression of human telomerase reverse transcriptase (hTERT), which encodes the catalytic and a rate-limiting subunit of telomerase. Adult human MSC cultured in vitro lack telomerase activity [28] and hTERT expression [29], which results in telomere shortening with serial passaging [29]. However, when hTERT is ectopically overexpressed, telomerase activity is restored resulting in longer telomeres and increased cellular lifespan [30]. Telomerization of MSC by hTERT overexpression maintains stem cell characteristics of MSC during long-term culture and allows large-scale cell expansion [30]. Some but not all early human fetal tissues express telomerase [31], but telomere and hTERT activities have not been investigated in fetal MSC.

With a view to understanding the relative benefits of fetal versus adult sources for MSC therapy, our goal was to explore whether first-trimester fetal MSC from blood, liver, and bone marrow were phenotypic intermediates between ESC and adult MSC. We characterized pluripotency markers, growth kinetics, and telomere status in human first-trimester fetal blood, liver, and bone marrow MSC and compared them with adult bone marrow MSC.

Materials and Methods


Blood and tissue collection for research purposes was approved by the Research Ethics Committee of Hammersmith and Queen Charlotte's Hospitals. National guidelines (Polkinghorne) were complied with in relation to the use of fetal tissues for research. Pregnant women gave written consent for the clinical procedure and for the use of blood or tissue for research purposes.

Cell Sources

Human first-trimester fetal MSC from blood, liver, and bone marrow were harvested as described previously [12]. Adult MSC from bone marrow were obtained from Prof. F. Dazzi (Imperial College London, London, U.K.) and Tulane Center for Gene Therapy (Tulane University, New Orleans, LA). Endometrial cells were obtained from Prof. J. Brosens (Imperial College London). TERT-MSC DNA and proteins were from Prof. M. Kassem (University of Aarhus, Aarhus, Denmark). The immortalized human embryonic kidney cell line 293T [32] and the human immortalized osteoblast cell line Saos-2 cells were purchased from ATCC (CRL-11268 and CRL HTB-85; Teddington, U.K., Human fetal MSC were characterized by their adherence to plastic, their fibroblast-like spindle morphology, and their immunophenotype. Consistent with their nonhematopoietic, nonendothelial origin, cultured human fetal MSC were all CD45-, CD34-, CD14-, and CD31-negative. Furthermore, we confirmed their mesenchymal nature by immunophenotyping (SH2-, SH3-, and SH4-positive and expression of laminin, fibronectin, and vimentin) and ability under permissive conditions to differentiate down the osteogenic and adipogenic lineages [12, 33].

Tissue Culture

All cells were plated at 104 cells per cm2 and cultured in 10-cm2 dishes with expansion medium (i.e., Dulbecco's modified Eagle's medium; Sigma-Aldrich, St. Louis, supplemented with 10% fetal bovine serum (Gibco-BRL, Gaithersburg, MD, and 2 mmol/l l-glutamine, 50 IU/ml penicillin, 50 mg/ml streptomycin (Gibco-BRL) at 37°C in 5% CO [2]. Subconfluent (70%–80%) cells were detached with 0.05% trypsin-0.01% EDTA (Gibco-BRL), and only early passage cells (passages 2–7) were studied.


Cells grown exponentially were fixed in 4% PFA in 125 mM HEPES (pH 7.6; 10 minutes, 4°C), 8% PFA in the same buffer (50 minutes, 4°C) and permeabilized in 0.5% Triton X-100 in phosphate-buffered saline (PBS) (30 minutes, gentle rocking). After fixation and permeabilization, cells were rinsed (3×) in PBS, incubated (30 minutes) with 20 μM glycine in PBS, blocked (1 hour) with PBS+ (PBS supplemented with 1% BSA, 0.2% fish skin gelatin, 0.1% casein; pH 7.6), incubated (2 hours) with primary antibodies in PBS+, washed (5× over 1.5 hours) in PBS+, incubated (1 hour) with secondary antibodies in PBS+, washed (overnight, 4°C) in PBS+, and rinsed (3×) in PBS, before being mounted in VectaShield labeled with 4,6-diamidino-2-phenylindole (Vector Laboratories, Burlingame, CA, and visualized immediately. For fluorescence microscopy (Axioscope I microscope equipped with CCD camera and iPlab software; Carl Zeiss, Jena, Germany,, images were collected sequentially (TIFF files), transferred to Adobe Photoshop (Adobe Systems Inc., San Jose, CA,, and contrast-stretched without further processing. The following primary antibodies were used: mouse monoclonal IgG Oct-4 (Santa Cruz Biotechnology Inc., Santa Cruz, CA,, IgG goat polyclonal Nanog (Santa Cruz Biotechnology), mouse monoclonal IgM SSEA-3, IgG SSSEA-4, IgM TRA-1–61, and IgM TRA-1–81 (from the ES Cell characterization kit; Chemicon, Temecula, CA, All primary antibodies were used at a 1:50 dilution. Secondary antibodies for immunofluorescence were donkey anti-mouse or anti-goat IgG conjugated with fluorescein isothiocyanate (1:100 and 1:1000, respectively; multiple-labeling grade; Jackson Immunoresearch Laboratories, West Grove, PA,

Reverse Transcription-Polymerase Chain Reaction

Total RNA was extracted from cells using the RNeasy Mini RNA isolation kit (Qiagen, Hilden, Germany, Total RNA was eluted from the mini columns with 50 μl of RNase-free water. The amount of total RNA isolated was quantified by optical density at 260 nm (OD260). Starting from 1 μg of total RNA, 20 μl of cDNA was synthetized using Pd(N)6 random hexamers (Amersham Biosciences, Piscataway, NJ, and 1 μl of 200 U M-MLV reverse transcriptase in the presence of dNTPs (Promega, Madison, WI, The reaction was performed for 10 minutes at 25°C, for 60 minutes at 42°C, and for 10 minutes at 75°C. cDNA was stored at −20°C until use.

cDNA was amplified by 30 cycles of denaturation (60 seconds at 94°C), annealing (30 seconds, 60°C), and elongation (30 seconds, 72°C), followed by a final step at 72°C for 5 minutes, using primer sequences previously published [34] and blasted against expected published sequences to confirm specificity (data not shown).

Growth Kinetics

To investigate whether fetal MSC were more readily expandable than adult MSC, we compared their growth kinetics, estimated by the cumulative population doubling over 50 days. Fetal and adult MSC were plated in triplicate at a concentration of 104 cells per cm2 in 10-cm2 dishes and successively subcultured at the same density when subconfluent [35]. The cells were detached and counted in a hemocytometer in trypan blue to exclude dead cells. This replating procedure was serially repeated over 50 days, and the cumulative cell doublings of the populations were plotted against time in culture to determine the growth kinetics of fetal and adult MSC expansion. The number of population doublings was determined by counting the number of adherent cells at the start and end of each passage.

To compare the growth rate of fetal and adult MSC at different cell density and to determine the effect of basic fibroblast growth factor (bFGF), we seeded bone marrow-derived adult MSC and fetal MSC from the same source (bone marrow) in triplicate in 12-well plates at 10, 100, 1,000, and 10,000 cells per cm2 with or without bFGF (5 ng/ml). The number of cells was counted every 2 days over a 2 weeks period, and the growth rates were determined by plotting the number of cells against time for different conditions. The relative cell morphology and size of fetal and adult bone marrow-derived MSC was assessed under a microscope (magnification, ×10 and ×20) on adherent cells stained by crystal violet and in trypsinized cells visualized in a hemocytometer chamber.

Real-Time Polymerase Chain Reaction Quantification

We used SYBR Green dye fluorescence (Applied BioSystems, Foster City, CA, and the ABI Prism 7700 Sequence Detection system (Applied Biosystems) based on the assumption that any increase in fluorescence signal is proportional to the amount of specific polymerase chain reaction (PCR) product. As the cycle threshold value (Ct) to reach fluorescence is directly proportional to the log of the initial amount of input cDNA, we estimated the amount of target sequence in the experimental sample by plotting the Ct of an unknown sample on a standard curve created with serial dilution of a reference sample. All samples were run at least in duplicate. Standard curves were generated with the ABI Prism 7700 SDS 1.7 software, and r2 values were consistently ≥0.998.

Relative Telomere Length Using Quantitative Real-Time PCR

Genomic DNA from experimental samples (Table 1) and the reference 293T cells was obtained using standard salting-out extraction after Miller et al. [36] and stored at −20°C until analysis. Relative telomere lengths were measured by SYBR Green real-time quantitative PCR amplification of telomere repeats (T) and single-copy gene 36B4 (S). The 36B4 gene was used to normalize for sample variations in DNA amount. The primers and thermal cycling profiles were adapted from Cawthon et al. [37]. T and S standard curves (Ct vs. log quantity) were generated using serial dilution of DNA (200 to 12.5 ng) from the telomerase-positive cell line 293T. Ct values in experimental samples were determined from semi-log amplification plots, and the standard curve was used to determine the quantity of telomere repeats, which is equivalent to the level of dilution of the arbitrary reference 293T needed to give the identical number of PCR cycles during the exponential phase. Relative telomere length was thus estimated as the Tquantity/Squantity (T/S) ratio, which is proportional to the mean terminal restriction fragment telomere (TRF) length [37]. The telomere and 36B4 PCRs were run on separate plates, and the standard curve was included in each run to allow relative quantification between samples (50 ng per sample). The concentration of reagents was 25 μl of 2× SYBR Green PCR Master Mix (Applied Biosystems), with the following final primer concentrations: Tel 1, 270 nM; Tel 2, 900 nM; 36B4u, 300 nM; and 36B4d, 500 nM. The thermal cycling conditions started with 95°C for 10 minutes; this step was followed by 40 cycles of 95°C for 15 seconds and then 54°C for 2 minutes for telomere PCR, and with 40 cycles of 95°C for 15 seconds and then 58°C for 1 minute for 36B4 PCR.

Table Table 1.. Samples
original image

Quantitative Real-Time PCR Relative Telomerase Activity

Relative telomerase activity was measured by a SYBR Green real-time quantitative telomeric repeat amplification protocol assay (TRAP) using a modification of the method of Wege et al. [38] In brief, cells were cultured to approximately 70% confluence and harvested by trypsinization. The resulting pellets were lysed using 1× CHAPS buffer, adjusted to 104 cells per microliter and frozen at −80°C until use. The concentration of reagents in the PCR amplification reaction was 12.5 μl of 2× SYBR Green PCR Master Mix (Applied Biosystems) with 0.1 μg of telomerase primer TS and 0.05 μg of anchored return primer ACX in a total volume of 25 μl. The thermal cycling conditions started with 25°C for 20 minutes followed by 35 cycles of 95°C for 30 seconds and 60°C for 90 seconds (two-step PCR). The sequences for telomerase primers (TS) and the anchored return primer ACX were as described by Kim and Wu [39]. The standard curve was generated using serial dilution of cell extracts from the 293T cell line. CT values in experimental samples were determined from semi-log amplification plots and the standard curve was used to determine the quantity of telomerase product; telomerase activity in experimental samples was then expressed as a percentage relative to calibrator 293T cells. We calculated in parallel the average ΔCT for each experimental sample, with ΔCT = CT(experimental sample) − CT(293T calibrator).

Real-Time PCR Quantification of hTERT mRNA

Levels of hTERT expression were estimated by SYBR Green real-time quantification using primers and methods described by Buttitta et al. [40] To normalize hTERT expression for intersample differences in RNA input, quality, and reverse transcriptase efficiency, we amplified the housekeeping gene β-actin. The ratio between copy numbers of hTERT mRNA and β-actin mRNA was used to normalize the amount of hTERT mRNA for each sample and allowed comparison between samples.

Statistical Analysis

After confirming normal distribution on histograms, standard parametric descriptive and two-tailed comparative statistics were used, and correlations were sought by linear regression using the least squares method. A value of p < .05 was considered significant. We used one-way analysis of variance for continuous variables.


Pluripotency Markers

Using immunofluorescence histochemistry, fetal populations homogenously expressed Oct-4, Nanog, SSEA-3, SSEA-4, Tra-1-61, and Tra-1-81 (Fig. 1A). Saos cells (human osteoblast cells) were used as negative control. In addition, because there is no antibody against Rex-1 epitope, and to confirm Oct-4 immunostaining, we determined Rex-1 and Oct-4 expression by reverse transcription (RT)-PCR, showing expression in all fetal but not adult MSC samples (Fig. 1B). Characterization of these cells confirmed that they were all capable to undergo osteogenic, adipogenic, and chondrogenic differentiation under permissive conditions, showing positive staining for calcium deposition (Alizarin red and von Kossa), adipose droplets (oil red O), and collagen (safronin red), respectively (Fig. 1C). Comparison of relative cell morphology and size of fetal and adult bone marrow-derived MSC cultured with or without bFGF for 1 week revealed that fetal MSC have a fibroblast-like morphology with spindle-shaped cytoplasm, whereas adult MSC have a larger cytoplasmic volume when cultured without bFGF. However, in the presence of bFGF, the morphology of adult MSC changed, with a reduction in cytoplasmic volume, also illustrated by the smaller size of trypsinized cells in the hemocytometer chamber. However, bFGF did not noticeably modify the morphology of fetal MSC (Fig. 1D) but triggered osteogenic differentiation, as seen by calcium deposition stained by von Kossa assay.

Figure Figure 1..

Expression of pluripotency markers Oct-4, Nanog, SSEA3, SSEA4, Tra-1-61, and Tra-1-81 by immunofluorescence (A) and of Oct-4 and Rex-1 by reverse transcription-polymerase chain reaction (B) in f MSC from blood, liver, and BM and Ad MSC. (C): Osteogenic, adipogenic, and chondrogenic differentiation in f blood, liver, and BM MSC, and Ad BM MSC, respectively visualized by von Kossa, oil red O, and safranin O staining. (D): Morphology of Ad and f BM-derived MSC stained with crystal violet and after trypsinization in an HC when cultured with or without bFGF. (E): Von Kossa staining of f and Ad BM MSC seeded at 10 000 cells per cm2 after 10 days in culture with or without bFGF. Abbreviations: Ad, adult; bFGF, basic fibroblast growth factor; BM, bone marrow; f, fetal; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; HC, hemocytometer.

Growth Kinetics

The comparative growth kinetics of fetal and adult MSC over 50 days at cell density of 10,000 cells per cm2 showed that fetal MSC had greater growth potential, achieving 28.4 population doublings in culture, whereas adult bone marrow MSC achieved only 7.1 (p < .01), with cumulative population doublings progressively increasing for fetal MSC but slowing in adult MSC (Fig. 2).

Figure Figure 2..

Growth kinetics.(A): Growth kinetics of MSC from human first-trimester fetal and Ad BM MSC seeded at 10,000 cells per cm2 estimated by the cumulative population doublings over 50 days. (B): Growth kinetics over 288 hours for fetal and Ad BM MSC seeded at various cell densities (10, 100, 1,000, and 10,000 cells per cm2) with or without bFGF (5 ng/ml), with the average doubling time (mean ± SD) indicated for each condition. Abbreviations: Ad, adult; bFGF, basic fibroblast growth factor; BM, bone marrow; f, fetal.

The analysis of growth rates when seeding cells at increasing density with or without bFGF revealed that cell density did not significantly affect the number of population doublings or the average doubling time for fetal and adult MSC, and in consequence, the highest numbers of cells were obtained from cultures seeded at 10,000 cells per cm2. For each seeding conditions, fetal MSC had higher growth rate than adult MSC (p < .01), and bFGF supplemented growth medium increased growth rate for both fetal and adult MSC in all conditions. However, adult MSC cultured in bFGF medium never reached the level of fetal MSC cultured without bFGF.

Telomere Length

Figure 3D shows that fetal MSC had longer telomeres than adult MSC (p < .001), whereas adult MSC and endometrial cells were similar in length (fetal blood, 74.6% of reference 293T value; fetal liver, 71.7%; fetal bone marrow, 75%; p < .001 compared with adult bone marrow and endometrial cells, 34%). Applying a conversion factor [37] to convert T/S ratios to their corresponding mean TRF length, fetal blood, liver, and bone marrow were 11, 10.8, and 11.2 kb, respectively, which corresponds to approximately ¾ of the length of hTERT-maximized MSC, whereas adult bone marrow MSC and endometrial cells were 7.2 and 7 kb, respectively.

Figure Figure 3..

Relative telomere length expressed as T/S ratios. The standard curves for T (A) and S (B) were obtained from serial dilutions of DNA (200 ng to 12.5 ng) from 293T cells, T/S ratios were plotted against passage number to show the distribution of the sample population and test for an effect of passage number on telomere ratios (C). The mean ± SEM relative telomere lengths (T/S ratio), expressed as a percentage of the T/S value in (D), were greater in F blood, liver, and BM than in Ad BM MSC, which did not differ significantly from the endometrial somatic cell population used as a low-level reference. Abbreviations: Ad, adult; BM, bone marrow; F, fetal; hMSC, human MSC; Rn, relative fluorescence; S, single-copy gene 36B4 amplification; T, telomere amplification.

Telomerase Activity and hTERT Gene Expression

Figure 4B shows a tight linear correlation between the two methods used (p < .01). Telomerase activity was not related to passage number over the range tested [2, [3], [4], [5], [6]–7] (Fig. 4C, 4D). Mean telomerase activity (expressed as a percent relative to 293T cells or as 2−ΔCT relative to 293T cells) was higher in fetal blood, liver, and bone marrow than adult bone marrow (p < .01). Adult MSC had repressed telomerase activity, similar to control adult endometrial cells (Fig. 4E). As expected, telomere length positively linearly correlated (p < .01) (Fig. 5). To know whether higher telomerase activity in the fetal samples corresponded to a greater amount of telomerase or higher enzymatic activity, we determined the quantities of hTERT mRNA relative to adult MSC using real-time RT-PCR and the ΔCT method (Fig. 6). hTERT mRNA was expressed in fetal blood, liver, and bone marrow 5.2, 4.8, and 5.7 times more, respectively, than in adult bone marrow samples.

Figure Figure 4..

Telomerase activity was calculated relative to the 293T cell line calibrator and expressed either as percentage quantity or 2−ΔCT. A standard curve was generated from serial dilution of 293T cell line lysates (A, B). Telomerase results expressed as percentage quantity relative to 293 or 2−ΔCT(C, D). Mean ± SEM telomerase activity relative to 293T cells using either method (E) showed greater activity in F samples compared with adult BM MSC and endometrial cells. Abbreviations: BM, bone marrow; F, fetal; hMSC, human MSC.

Figure Figure 5..

Correlation between telomere length (y axis) and telomerase activity (x axis). ∗, p < .01.

Figure Figure 6..

Quantity of human TERT RNA (mean + SEM) relative to Ad MSC using the ΔCT method. Abbreviations: Ad, adult; BM, bone marrow; f, fetal.


This study establishes that fetal MSC from human first-trimester blood, liver, and bone marrow have several advantageous biological characteristics over their adult bone marrow-derived counterparts, of downstream relevance to tissue engineering and cell therapy. Undifferentiated first-trimester fetal MSC, in contrast to adult MSC, expressed the pluripotency markers Oct-4, Nanog, Rex-1, SSEA-3, SSEA-4, Tra-1-60, and Tra-1-81 at the protein level. Fetal MSC also grew more rapidly with more population doublings than adult MSC, rendering them more expandable, with optimal cell density for cell expansion ranging from 1,000 to 10,000. Although the addition of bFGF in the expansion medium increases growth rate, it also modifies cell phenotype, reducing the size of the cells for adult MSC and triggering osteogenic differentiation for fetal MSC. This result is in line with the recent work of Solchaga et al., who showed that bFGF enhances both the mitotic and chondrogenic potential of adult bone marrow-derived MSC [41]. We also showed that fetal MSC have longer telomeres and higher telomerase activity than adult MSC. These data suggest that fetal MSC may prove preferable for cell/gene therapy and tissue regeneration. In keeping with this, human fetal MSC have recently been shown to express more primitive developmental genes than adult MSC, as well as genes implicated in their increased growth potential [13].

Although pluripotency marker expression has been considered a hallmark of ESC, this study shows that all three fetal sources of MSC express pluripotency markers. Although no previous study has tested this in fetal MSC, our finding is not entirely unexpected, given their earlier developmental origins and the other differences we have shown in fetal versus adult MSC. A few reports have suggested that early human MSC may express some pluripotency markers. Some MSC from second-trimester amniotic fluid express Oct-4 [9, 18], whereas Karlmark et al. [42] recently showed that second-trimester human amniotic fluid contains cell populations capable of activating Oct-4 and Rex-1 promoters. In addition, both hemopoietic and nonhemopoietic stem cell populations from umbilical cord blood have been shown to express some pluripotency markers, although these cells were not considered mesenchymal in origin [43, [44]–45].

We have previously shown multilineage potential of fetal MSC cultures, in particular that fetal but not adult MSC readily undergo myogenic differentiation [33]. Compared with adult bone marrow MSC, we recently found that the osteogenic efficiency of human fetal MSC was greater than adult MSC both in vitro and in vivo (unpublished data). Similarly in the rhesus monkey, third-trimester fetal MSC differentiated more readily down the osteogenic, chondrogenic, and adipogenic lineages than adult bone marrow-derived MSC [11]. Banfi et al. [35] showed that adult bone marrow MSC gradually lost their multiple differentiation potential during in vitro expansion, with their bone-forming efficiency in vivo reduced manyfold at first confluence compared with fresh bone marrow. Therefore, we speculate that lack of expression of early pluripotency markers in adult MSC may reflect a degree of multilineage lineage restriction.

Our results are comparable to previous reports documenting the slow replication and rapid senescence of adult MSC in vitro. The maximal population doublings we achieved in vitro were 70–80 for human fetal MSC [12], considerably greater than the 30–40 for adult MSC [35]. Adult MSC are only found at low frequency in bone marrow in vivo, and their low proliferative capacity is accompanied by progressive replicative aging and decrease in osteogenic potential when expanded in vitro, with no detectable telomerase activity [46]. Fetal MSC are smaller than adult MSC. Interestingly, cell morphology is affected by senescence, cell size increasing with in vitro ageing [24], whereas cells transduced with telomerase become smaller than their nontransduced counterparts [47].

Telomere length is relevant to cell therapy, especially in utero transplantation, when the recipient is younger, and stem cells will need to persist for a similar ontological lifespan. Telomeres protect chromosomal ends, and when protection is lost, certain genes can be activated to trigger senescence-related dysfunction and pathology, as in dyskeratosis congenital when DKC1 activation induces premature ageing and skin cancer [48]. Telomere status and cell proliferation are closely linked in MSC, with human adult MSC telomerized by overexpressing hTERT exhibiting extensive proliferation and acquiring an extended cell lifespan [49].

The potential therapeutic advantages of fetal MSC over adult MSC are not restricted to differentiation potential, growth kinetics, and telomere status. When transplanted into the fetus of immunodeficient SCID mice, donor cells from fetal liver cells showed a 10-fold engraftment advantage over those from adult bone marrow [50]. In preimmune fetal sheep, fetal MSC, like adult MSC, engrafted in multiple tissues and showed multilineage differentiation like their adult counterparts, but unlike adult bone marrow MSC, they appeared to contribute hemopoiesis [51]. In complementary work, we recently found that fetal MSC express a unique pattern of adhesion molecule expression, with higher levels of α2 and α4β1 and binding to their extracellular matrix ligands compared with adult MSC, which would facilitate homing and engraftment [52]. Fetal MSC also present some immunological advantages. Their lack of intracellular HLA class II and slower response to stimulation by γ-interferon [53] might confer fetal donor cells with an immune tolerance advantage.

Although early fetal MSC have been isolated from other fetal organs [54], human fetal MSC from blood liver and bone marrow are the best characterized. Allogeneic use raises the ethical issues associated with tissues derived from termination of pregnancy. Autologous use for ex vivo gene therapy, however, would allow collection of fetal blood and possibly liver in continuing pregnancies, and recent reports of MSC in first-trimester placenta [55] and mid-trimester amniotic fluid raise the possibility of using more accessible sources of fetal MSC.


We conclude that first-trimester human fetal MSC represent a developmentally less mature population of stem cells than adult MSC, with a number of characteristics advantageous to the development of cell therapy; these include expression of pluripotency markers, greater proliferative capacity, and longer telomeres maintained over passaging by higher levels of telomerase activity.


The authors indicate no potential conflicts of interest.


P.V.G. was supported by an Action Medical Research project grant, and C.G. was supported by the Medical Research Council, with additional infrastructural support from the Institute of Obstetrics & Gynaecology Trust. We thank Prof. M. Kassem for the DNA and protein from human MSC-TERT, Prof F. Dazzi for the adult bone marrow MSC, and Prof. J. Brosens for the endometrial cells.