Article

623

IntroductionThe first attempt at isolation of embryonic stem (ES) cells goesback to the 1960s, when Cole, Edwards and Paul (1965) obtainedES cells from rabbit embryos and compared the efficiency of theestablishment of ES cells from morula and blastocyst (seeEdwards, 2004). Even after this work, the major studies in thisfield involved for a long time only murine and humanteratocarcinomas or EC cell lines (Stevens, 1970), before ES cellswere established from the outgrowth of the inner cell mass (ICM)of murine blastocysts (Evans and Kaufman, 1981; Martin, 1981).These cells were shown to be positive for alkaline phosphatase(AP) and expressed stage-specific embryonic antigens SSEA-1,SSEA-3 and SSEA-4. The murine ES cell lines have also beenestablished from morula-stage embryo and were shown to besimilar to ES cells isolated from blastocyst (Eistetter, 1989). Themorula-derived ES cell lines were also obtained from mink andcow embryos (Sukoyan et al., 1993; Stice et al., 1996).

These developments have led to attempts to produce human EScells from an entire blastocyst (Bongso et al., 1994), and, finally,from ICM (Thomson et al., 1998). However, as yet there are noreports on the establishment of ES cells from human morula. This

paper describes the original technique for the derivation of ES celllines from human morula, which resulted in the establishment ofeight morula-derived ES cell lines, described here in comparisonwith the ES cell lines derived from the ICM and whole blastocyst.

Materials and methodsHuman embryos for the study were obtained from IVF andpreimplantation genetic diagnosis (PGD) programmes of theReproductive Genetics Institute (RGI) in different countries. Thestudy protocol, approved by the Institutional Review Board (IRB),allowed patients donating spare embryos for research by signingthe consent form. Overall, 117 embryos were obtained, including46 morula and 71 blastocyst-stage embryos (see Table 1).

Establishment of ES cell lines from humanmorula

Forty-six attempts were made to establish ES cell lines from themorula-stage embryos, which were cultured in medium G1 andthen in human tubal fluid (HTF) medium supplemented withplasmanate. The embryos were mostly normal diploids (seeTable 2, right column below). These embryos were obtained on

Morula-derived human embryonic stem cellsDr Strelchenko has over 20 years’ experience in cell culture, nuclear transfer technology,somatic cell hybridization and the establishment of embryonic stem cell (ESC) andprimordial germ cell lines. He has been principal investigator on cloning projects andcreated the world’s first cloned calf and numerous high genetic merit animals. At theReproductive Genetics Institute, he is director of the ESC research laboratory andparticipates in establishing and characterizing the human ESC lines repository, whichcurrently includes over 100 human ESC lines.

Nick Strelchenko1, Oleg Verlinsky, Valeri Kukharenko, Yury VerlinskyReproductive Genetics Institute, Chicago, IL, USA1Correspondence: e-mail: rgi@flash.net

Dr Nick Strelchenko

AbstractHuman embryonic stem (ES) cells are known to derive from the inner cell mass of blastocyst. Although the embryos of otherdevelopmental stages have also been used as a source for ES cells in animal models, the feasibility of obtaining ES cell linesfrom human morula is not known, despite being an obvious source available through assisted reproduction and preimplantationgenetic diagnosis programmes. This study describes an original technique for derivation of ES cells from human morula, whichenabled the establishment of eight morula-derived ES cell lines. These ES cell lines were shown to have no morphologicaldifferences from the ES cells derived from blastocysts, and expressed the same ES cell specific markers, including Oct-4,tumour-resistance antigens TRA-2–39, stage-specific embryonic antigens SSEA-3 and SSEA-4, and high molecular weightglycoproteins TRA-1–60 and TRA-1–81, detected in the same colony of morula-derived ES cells showing specific alkalinephosphatase expression. No differences were observed in these marker expressions in the morula-derived ES cells cultured inthe feeder layer free medium. Similar to ES cell originating from blastocyst, the morula-derived ES cells were shown tospontaneously differentiate in vitro into a variety of cell types, including the neuron-like and contracting primitive cardiocyte-like cells.

Keywords: blastocyst, embryonic stem (ES) cells, ES cell marker expression, human ES cell lines, morula

RBMOnline - Vol 9. No 6. 2004 623-629 Reproductive BioMedicine Online; www.rbmonline.com/Article/1558 on web 27 October 2004

624

Article - Morula-derived human embryonic stem cells - N Strelchenko et al.

day 4, when PGD had been established. Otherwise the embryoswere frozen, especially when they were transported fromdifferent PGD centres. The technique involved the removal ofzona pellucida by pronase (3 mg/ml; Sigma, St Louis, MO,USA), injection of the naked morula by use of micromanipulatorunder feeder layer, which was the mitotically inactivated murineprimary embryonic fibroblasts or BRL cells, inactivated bymitomycin C (Sigma) at 10 μg/ml for 5–6 h. The cells werecultured in Alpha MEM or DMEM medium, supplemented with10–20% of fetal bovine serum (FBS) or knockout serumreplacement (SR-1) (all from Gibco, Grand Island, New York,USA). In addition, beta-mercaptoethanol (Gibco) at a finalconcentration of 1 mmol/l and human recombinant basicfibroblast growth factor (FGF; Sigma) at a final concentration of5 ng/ml were added to culture medium, and after cell outgrowthand spreading into the layer, which was observed overapproximately 8–14 days, the initial disaggregation wasperformed (passage 0) using 1 mmol/l EDTA (Sigma) in Hank’sor PBS Ca2+–Mg2+-free solutions. Using glass needles, only softloose cell clumps were cut and transferred into a new dish with afeeder layer, while loose cells were carefully transferred onto the

feeder layer and proliferate. Fast proliferating colonies with ES-like morphology were isolated and propagated further. Within thenext two to five passages, the uniform proliferating cells wereselected, and colonies of established ES cell lines were passagedusing collagenase V (Sigma) at 1.5 μg/ml in HTF–HEPES(Gibco) or 1 mmol/l EDTA, followed by harvesting with a celllifter (Costar, Corning, New York, USA); the undifferentiated EScell population was then isolated using EDTA solution (thepatent for morula-derived ES cell lines is pending). On average,there were 4–6 population doublings per passage.

Establishment of ES cell lines from ICM ofhuman blastocyst

Of 71 attempts undertaken for establishment of ES cell linesfrom blastocyst-stage human embryos, blastocystimmunosurgery was performed (Thomson et al., 1995) in 32 ofthem to isolate the ICM cluster, using 1:50 diluted humanantiserum Sigma (H3383) and 1:3 diluted guinea pig serumcomplement (Sigma, S1639) in HTF–HEPES. After mechanicalremoval of collapsed trophoblast, the isolated ICM (or the whole

Table 1. Efficiency of establishing ES cells from different stage embryos.

Embryo stage Number Outgrowth Cell lines % (n) (n) establishment

Morula 46 11 8 17Blastocyst 39 12 7 18ICM 32 5 5 16

Table 2. Characterization of ES cell lines derived from different stage embryos.

Cell Source No. AP SS3 SS4 T60 T80 T39 Oct4 KaryotypeLine passages

15 Morula 17 100 87 95 97 98 100 94 46, XY18 Morula 16 100 91 96 95 96 100 96 46, XX21 Morula 14 100 + + + + 100 91 46, XX24 Morula 10 100 + + + + 100 + 46, XX27 Morula 15 100 77 + + + 100 + 46, XX28 Morula 16 100 69 89 93 91 100 97 46, XY31 Morula 10 100 72 + + + 100 + 46, XX33 Morula 8 100 84 + + + 100 + 46, XY53 Blast 6 100 + + + + 100 + 46, XX60 Blast 5 100 84 93 97 98 100 94 46, XX62 Blast 7 100 + + + + 100 + 46, XY63 Blast 6 100 76 92 94 92 100 94 46, XY79 Blast 7 100 + + + + 100 + 46, XX80 Blast 24 100 + + + + 100 + 46, XX81 Blast 6 100 + + + + 100 + 46, XY93 ICM 5 100 76 91 95 99 100 96 46, XY94 ICM 5 100 + + + + 100 + 46, XY95 ICM 4 100 + + + + 100 + 46, XX96 ICM 21 100 + + + + 100 + 46, XX97 ICM 3 100 83 96 97 98 100 95 46, XY

Abbreviations: ICM = inner cell mass, blast = blastocyst, AP = alkaline phosphatase, SS3 = SSEA-3 (stage-specific embryonic antigen 3), SS4 = SSEA-4 (stage-specific embryonic antigen 4), T60 = TRA-1–60 (tumour resistance antigen-1-60), T80 = TRA-1–80(tumour-resistance antigen-1-80), T39 = TRA-2–39 (tumour-resistance antigen-2-39), Oct4 = Oct4 gene (belongs to the POU family oftranscription factors).

625

Article - Morula-derived human embryonic stem cells - N Strelchenko et al.

blastocyst) was placed on the top of feeder layer, withproliferating cells observed within 24–48 h. Passage 0 of theblastocyst outgrowth cells was performed using 1 mmol/l EDTAin PBS or in Hank’s Ca–Mg-free solution. The cell colonies wereselected 7–10 min after exposure to 1 mmol/l EDTA inphosphate-buffered saline or in Hank’s Ca–Mg-free solution,with further passages performed by collagenase or 1 mmol/lEDTA in Hank’s Ca–Mg-free solution, followed by harvestingusing the cell lifter. The cell suspension was put through the 100 mkm cell strainer before placing on the fresh feeder layer.

ES cell marker expression

Alkaline phosphatase was detected by SK-5300 kit (VectorLaboratories, Inc., Burlingame, CA, USA), and by primaryantibodies TRA-2–39 isotype IgG1 (Santa Cruz Biotechnologies,Inc., Santa Cruz, CA, USA; cat. no. SC-21708), developedagainst epitope 2102 of L-alkaline phosphatase of humancarcinoma, using goat anti-mouse isotype IgG1 labelled withfluorescein isothiocyanate (FITC; Nikon, filter B-1A).

Monoclonal rat antibodies were used for detection of the stage-specific antigen SSEA-3 (Santa Cruz Biotechnologies, cat. no.SC-21703). In addition, monoclonal rat antibodies were used forthe detection of isotype IgM (Jackson ImmunoResearchLaboratories, Inc., West Grove, PA, USA; cat. no. 112–095–075;Isotype IgM labelled with FITC). Detection of SSEA-4 wasperformed with primary mouse monoclonal antibodies (IsotypeIgG3; Santa Cruz Biotechnologies, cat. no. SC-21704 combinedwith secondary FITC-labelled antibodies).

High molecular weight glycoproteins or tumour rejectionantigens, TRA-1–60, were detected with primary mousemonoclonal antibodies (Santa Cruz Biotechnologies, cat. no. SC-21705) and secondary antibodies SC-2082 (isotype IgM labelledwith FITC). Glycoprotein antigen detected by TRA-1–80 wasdetected with primary antibodies (isotype IgM labelled withFITC; Santa Cruz Biotechnologies, cat. no. SC-21706).

Localization of Oct-4 expression in ES cells was performed withpolyclonal antibodies (Santa Cruz Biotechnology) and secondarylabelled with TRITC, and Oct-4 expression was assessed byGene Choice One Tube reverse transcriptase-polymerase chainreaction (RT-PCR) kit (Vector Laboratories, Inc.), using primersequences CACGAGATGCAAAGCAGAAACCCTCGG andTTGCCTCTCACTCGGTTCTCG, generating the RT-PCRproduct of 73 bp.

The same markers were tested in ES cell cultures in the feederlayer free medium (Carpenter et al., 2004; Rosler et al., 2004).

Differentiation of ES cells in vitro

Spontaneous differentiation of ES cells into a variety ofdifferentiated cell types was studies following in-vitro initiationof embryonic bodies from ES cell lines obtained from embryosfrom different stages of development (Reubinoff et al., 2000).

Chromosomal analysis

Chromosomal analysis of the established ES cell lines was doneusing standard karyotyping by G-banding before freezing.

Freezing and storage

The established ES cell lines were tested for mycoplasmal andbacterial contamination and frozen in Alpha MEM mediumcontaining 5% DMSO, 5% glycerol and 10% FBS or 15% SR1,using the control rate freezer, lowered temperature for at a rate of1°C per min to –30°C, followed by storage of cells in vapour ofliquid nitrogen (–176°C).

ResultsFigure 1 shows the different steps in the derivation of ES cellsfrom the human morula, described in Materials and methodsabove, in comparison with ES cell from ICM. After placing thenaked morula under the confluent feeder layer, cell proliferationwas observed, creating a plate of compacted cells (Figure 1a),which further proliferated in the culture dish as shown in Figure1b. Following the initial passage of proliferating cell clusters, thedisaggregated cells were placed on the top of the feeder layer,with cell outgrowth observed in eight of 11 morula culturesinitiated. During the next two to four passages, only cell colonieswith morphology of ES cells were selected for a further culture.It was at this stage when three of 11 morula-derived cell linesfailed to further proliferate, with the remaining eight resulting inthe stable and monomorphic ES cell lines, as shown in Figure1c. Overall, only 11 (23.9%) of 46 morula-stage embryosresulted in the initial outgrowth, with eight (17%) yielding EScell lines, similar to the outcome of ES cell line originating fromblastocyst (Table 1).

In contrast to morula-derived ES cell lines, derivation of ES cellsfrom entire blastocysts required precise timing for initialdisaggregation (passage 0). Although initial outgrowth wasobserved in 12 (30.7%) of 39 initiated cultures originating fromthe entire blastocysts, five of them were lost in the selectionprocess during passages 2 and 3, resulting in the establishment ofES cell lines in only seven (18%) of 39 attempts.

Similar results were obtained from ICM, although this involveda more time-consuming immunosurgery technique. Theprocedure of the derivation of ES cell line from ICM is shown inthe right column of Figure 1, starting with the attachment ofICM to feeder layer (Figure 1d), initial outgrowth (Figure 1e)and the establishment of the proliferation of ES cells (Figure 1f),without selection required in ICM-derived ES cell lines.Although only five (15.6%) of 32 attempted ICM culturesproduced initial outgrowth after introduction in vitro, all of themresulted in the establishment of ES cell lines.

As shown in Table 2, all 20 established ES cell lines expressedspecific ES stem cell markers, including AP, SSEA-3, SSEA-4,TRA-1–60, TRA-1–80, TRA-2–39 and Oct-4, irrespective oftheir origin (Figure 2c,d – SSEA-3; 2e,f – TRA-1-80). Thesemarkers were detected in the same colonies for morula,blastocyst and ICM-derived cell lines, expressing AP (see Figure2), such as immunostaining with primary TRA-2–39 andsecondary FITC-labelled antibodies in morula (Figure 2a) andICM-derived cell lines (Figure 2b), as well as red fluorescence(TRITC) for Oct-4 (Figure 2a,b) for morula and ICM-derivedES cells respectively. There were no differences in the aboveexpression patterns following culture of ES cells of differentorigin in the feeder layer free medium.

626

Article - Morula-derived human embryonic stem cells - N Strelchenko et al.

Figure 1. Initial steps in establishing of human ES cell lines from different embryo stages. (a) Placing a human embryo at morulastage under mouse embryonic fibroblasts for establishing ES cells (×10, Hoffman DIC). (b) Initial growth of morula cells underfeeder layer (×10, Hoffman DIC). (c) Morphology of human morula-derived ES cells (×10, Ph1); (d,e) Initial growth of ES cellsout of isolated ICM (×20, Ph1). (f) Primary colony of ES cells derived from ICM (×20, Ph1). Morphology of human ICM-derivedES cells (×10, Ph1).

a d

b e

c f

627

Article - Morula-derived human embryonic stem cells - N Strelchenko et al.

Figure 2. Expression of markers TRA-2–39, Oct-4, SSEA-3, TRA-1–80 and their match on the same colonies with enzymatic activityof alkaline phosphatase (AP) in human embryo stem (ES) cell lines derived from morula (left) and inner cell mass (ICM, right). [TRA= tumour-resistance antigen; SSEA = stage-specific embryonic antigen; FITC = fluorescein isothiocyanate; TRITC =tetramethylrhodamine isothiocyanate.] (a) Immunofluorescence of TRA-2–39 (FITC) and Oct-4 (TRITC) of human ES-cells derivedfrom morula and growing on the murine feeder layer stained by Hoechst 33342. (Objective 20×, filter Nikon D/F/T.) (b) Immunofluorescence of TRA-2–39 (FITC) and Oct-4 (TRITC) of human ES-cells derived from ICM and growing on the murinefeeder layer stained by Hoechst 33342. (Objective 20×, filter Nikon D/F/T.) (c) Expression of enzymatic activity of AP and matchedimmunofluorescence cell surface staining of SSEA-3 detected by monoclonal antibodies labelled with FITC on the same colony ofhuman ES cells derived from morula and growing on the murine feeder layer. (Objective 20×, filter Nikon B-1A.) (d) Expression of enzymatic activity of AP and matched immunofluorescence cell surface staining of SSEA-3 detected by monoclonalantibodies labelled with FITC on the same colony of human ES cells derived from ICM and growing on the murine feeder layer.(Objective 20×, filter Nikon B-1A.) (e) Expression of enzymatic activity of AP and matched immunofluorescence cell surface stainingof TRA-1–80 detected by monoclonal antibodies labelled with FITC on the same colony of human ES cells derived from morula andgrowing on the murine feeder layer. (Objective 20×, filter Nikon B-1A.) (f) Expression of enzymatic activity of AP and matchedimmunofluorescence cell surface staining of TRA-1–80 detected by monoclonal antibodies labelled with FITC on the same colony ofhuman ES cells derived from ICM and growing on the murine feeder layer. (Objective 20×, filter Nikon B-1A.)

a b

c d

e f

628

Article - Morula-derived human embryonic stem cells - N Strelchenko et al.

Three morula-derived ES cell lines were with normal male(46,XY) and five normal female (46,XX) karyotypes. Similarly,three blastocyst-derived ES cell lines were with normal male,and four with female karyotypes, while three ICM-derived EScell lines were with normal male and two with normal femalekaryotypes.

After in-vitro initiation of embryonic bodies from ES cell linesoriginating from different stages of embryo development, a widevariety of differentiated cells were obtained, including neuronsand primitive cardiocytes. All established ES cell lines werefrozen with control thaw-out.

DiscussionThese data present the first evidence for the establishment of EScell lines from the human morula. This is despite the fact thatmorula cells are significantly different from blastocyst cells, notonly in size of adjacent cytoplast but also in gene patternexpression. This includes an unstable maternal methylationpattern that persists until the morula stage, disappearing at theblastocyst stage, where low concentrations of methylation arepresent independently from parental origin (Hanel and Wevrick,2001). For example, bovine embryos display a high sensitivityto ouabain (potent inhibitor of the Na/K-ATPase), withenzymatic activity undergoing a nine-fold increase from themorula to the blastocyst stage (Watson and Barcroft, 2001).mRNA expression patterns have been shown to be different inmouse embryos at the morula, and blastocyst stages (Lee et al.,2001; Tanaka and Ko, 2004).

The method of establishment of the morula-derived ES cell linesdescribed is simple and results in 17% efficiency of theestablishment of the human morula-derived ES cell lines,similar to the establishment of ES cell lines from the wholeblastocyst (18%) and ICM (16%). Despite the above differencesbetween morula and blastocyst, comparative studies of 20 EScell lines obtained in this study from the embryos of differentstages of embryonic development revealed no morphologicaldifferences between human ES cells originating from the entireblastocyst, from ICM and from morula. No differences wereobserved in the pattern of marker expression either, includingalkaline phosphatase (AP), TRITC-immunofluorescence ofexpression of Oct-4, immunofluorescence of TRA-2–39 (L-AP), TRA-1–60, and TRA-1–80, detected by monoclonalantibodies labelled by FITC, antigens SSEA-3 and SSEA-4.These patterns of marker expression in morula-derived ES cellswere also observed following the culture in feeder layer freemedium, and the cells were characterized by the same ability todifferentiate in different type of cells irrespective of their origin.However, further studies will be needed to investigate if morula-derived ES cells would display differential characteristics atlater stages of differentiation, depending on a possiblepartitioning of inside and outside cells of morula at the time ofharvesting, from which ES cells were derived. Because insidecells of morula were shown to express Oct-4 (Liu et al., 2004),the resulting morula-derived ES cells may have originated fromthese cells. The same cells may be involved in forming theepiblast of the blastocyst, from which ES cells have recentlybeen derived (Buehr and Smith, 2003).

Thus the ES cell lines may be derived from human morula,representing typical morphology and marker expression of ES

cells, which may be used in future research in thedevelopments and application of cell therapies.

Acknowledgement

The authors would like to express their gratitude to ProfessorAnver Kuliev for editing the paper and critical review of theresults of this work.

References

Bongso A, Fong CY, Ng SC, Ratnam S 1994 Isolation and culture ofinner cell mass cells from human blastocysts. HumanReproduction 9, 2110–2117.

Buehr M, Smith A 2003 Genesis of embryonic stem cells.Philosophical Transactions of the Royal Society of London, SeriesB, Biological sciences 358, 1397–1402.

Carpenter MK, Rosler ES, Fisk GJ et al. 2004 Properties of fourhuman embryonic stem cell lines maintained in a feeder-freeculture system. Developmental Dynamics 229, 243–258.

Cole RJ, Edwards RG, Paul J 1965 Cytodifferentiation in cellcolonies and cell strains derived from cleaving ova andblastocysts of the rabbit. Experimental Cell Research 37,501–504.

Edwards RG 2004 Stem cells today: A. Origin and potential ofembryo stem cells. Reproductive BioMedicine Online 8, 275–306.

Eistetter HR 1989 Pluripotent embryonal stem cells can beestablished from disaggregated mouse morulae. Development,Growth and Differentiation 31, 275–282.

Evans MJ, Kaufman MN 1981 Establishment in culture ofpluripotential cells from mouse embryos. Nature 292, 154–156.

Hanel ML, Wevrick R 2001 Establishment and maintenance of DNAmethilation patterns in mouse Ndn: implications for maintenanceof imprinting in target genes of the imprinting center. MolecularCell Biology 21, 2384–2392.

Lee KF, Chow JF, Xu JS et al. 2001 A comparative study of geneexpression in murine embryos developed in vivo, cutured in vitro,and cocultured with human oviductal cells using messengerribonucleic acid differential display. Biology of Reproduction 64,910–917.

Liu L, Czerwiec E, Keefe D 2004 Effect of ploidy and parentalgenome composition on expression of Oct-4 protein in mouseembryos. Gene Expression Patterns 4, 433–441.

Martin GR 1981 Isolation of a pluripotent cell line from early mouseembryos cultured in medium conditioned by teratocarcinomastem cells. Proceedings of the National Academy of Sciences ofthe USA 78, 7634–7638.

Reubinoff B, Pera MF, Fong CY et al. 2000 Embryonic stem celllines from human blastocysts: somatic differentiation in vitro.Nature Biotechnology 18, 399–404.

Rosler ES, Fisk GJ, Ares X et al. 2004 Long-term culture of humanembryonic stem cells in feeder-free conditions. DevelopmentalDynamics 229, 259–274.

Stevens LC 1970 The development of transplantableteratocarcinomas from intratesticular grafts of pre- andpostimplantation mouse embryos. Developmental Biology 21,364–382.

Stice SL, Strelchenko NS, Keefer CL, Matthews L 1996 Pluripotentbovine embryonic cell lines direct embryonic developmentfollowing nuclear transfer. Biology of Reproduction 54, 100–110.

Sukoyan MA, Vatolin SY, Golubitsa AN et al. 1993 Embryonic stemcells derived from morulae, inner cell mass, and blastocysts ofmink: comparisons of their pluripotencies. MolecularReproduction and Development 36, 148–158.

Tanaka TS, Ko MSH 2004 A global view of gene expression in thepreimplantation mouse embryo: morula versus blastocyst.European Journal of Obstetrics, Gynecogy and ReproductiveBiology 115S, S85–S91.

Thomson J, Kalishman J, Golos TG et al. 1995 Isolation of a primate

Article - Morula-derived human embryonic stem cells - N Strelchenko et al.

embryonic stem cell line. Proceedings of the National Academyof Sciences of the USA 92, 7844–7848.

Thomson JA, Itskovitz-Eldor J, Shapiro SS et al. 1998 Embryonicstem cell lines derived from human blastocysts. Science 282,1145–1147.

Watson AJ, Barcroft LC 2001 Regulation of blastocyst formation.Frontiers in Bioscience 6, 708–730.

Received 23 September 2004; refereed 5 October 2004;accepted 15 October 2004.

629