Oct-4-expressing cells in human amniotic ¯uid: a newsource for stem cell research?

Andrea-Romana Prusa, Erika Marton, Margit Rosner, Gerhard Bernaschek andMarkus HengstschlaÈger1

Obstetrics and Gynecology, University of Vienna, Prenatal Diagnosis and Therapy, WaÈhringer GuÈrtel 18±20, 1090 Vienna, Austria

1To whom correspondence should be addressed. E-mail: markus.hengstschlaeger@akh-wien.ac.at

BACKGROUND: It is the hope of investigators and patients alike that in future the isolation of pluripotent human

stem cells will allow the establishment of therapeutic concepts for a wide variety of diseases. A major aim in this

respect is the identi®cation of new sources for pluripotent stem cells. Oct-4 is a marker for pluripotent human stem

cells so far known to be expressed in embryonal carcinoma cells, embryonic stem cells and embryonic germ cells.

METHODS: Cells from human amniotic ¯uid samples were analysed for mRNA expression of Oct-4, stem cell

factor, vimentin and alkaline phosphatase via RT±PCR. Oct-4 protein expression was investigated by Western blot

analysis and immunocytochemistry. Oct-4-positive cells were also analysed for the expression of cyclin A protein via

double immunostaining. RESULTS: Performing RT±PCR, Western blot and immunocytochemical analyses

revealed that in human amniotic ¯uid in the background of Oct-4-negative cells a distinct population of cells can be

found, which express Oct-4 in the nucleus. Oct-4-positive amniotic ¯uid cell samples also express stem cell factor,

vimentin and alkaline phosphatase mRNA. The Oct-4-positive amniotic ¯uid cells are actively dividing, proven by

the detection of cyclin A expression. CONCLUSIONS: The results presented here suggest that human amniotic ¯uid

may represent a new source for the isolation of human Oct-4-positive stem cells without raising the ethical concerns

associated with human embryonic research.

Key words: human amniotic ¯uid/Oct-4/pluripotency/stem cells

Introduction

The Oct-4 POU transcription factor is expressed in mouse

totipotent embryonic stem and germ cells. The expression of

human Oct-4 is comparable with the pattern in the mouse,

suggesting that it may have a similar function in preventing

human totipotent embryo cells from differentiating. Each type

of established mammalian pluripotent stem cell lineÐembry-

onal carcinoma (EC) cells, embryonic stem (ES) cells and

embryonic germ (EG) cellsÐexpress Oct-4, which disappears

rapidly when the cells differentiate. Accordingly, Oct-4 is

considered to be a marker of pluripotent stem cells (Donovan,

2001; Pesce and SchoÈler, 2001).

Human amniotic ¯uid cells are used for routine prenatal

genetic diagnosis of a wide range of fetal abnormalities caused

by genetic alterations. Although this routine diagnosis is well

established, little is known about the origin and properties of

these cells. Cells of different embryonic/fetal origins of all

three germ layers have been reported to exist in amniotic ¯uid,

and for speci®c subsets of amniotic ¯uid cells the origins still

need to be clari®ed (Milunsky, 1979; Hoehn and Salk, 1982;

Gosden, 1983; Prusa and HengstschlaÈger, 2002).

The purpose of this study was to investigate whether human

amniotic ¯uid contains cells expressing the transcription factor

Oct-4.

Materials and methods

Amniotic ¯uid cells

Amniotic ¯uid cell samples were obtained from amniocentesis all

performed after the 14th week of pregnancy for routine prenatal

diagnosis. The indications were advanced maternal age and the

cytogenetic analyses revealed normal karyotypes. This project has

been reviewed and accepted by the ethics commission of the

University of Vienna, Austria (project number: 036/2002) and each

patient signed a written informed consent.

RT±PCR

RNA of amniotic ¯uid cells was prepared using TriReagent

(Molecular Research Center, Inc., Cincinnati, OH, USA). First-strand

cDNA synthesis and PCR were performed using the rp-RT Mix

(ViennaLab, Austria) following the protocol provided by the

manufacturer. Sequences and PCR conditions have been described

previously for the ®rst used Oct-4-speci®c oligonucleotide set (Monk

and Holding, 2001; in Figure 1 designated `primer set 1') and for the

second set (Xu et al., 2001; in Figure 1 designated `primer set 2'). The

oligonucleotides speci®c for stem cell factor have been designed and

synthesized by VBC Genomics (Vienna, Austria) according to the

sequence published by Martin et al. (1990), and the PCR conditions

were the same. For the analysis of vimentin mRNA expression,

speci®c oligonucleotides and PCR conditions were chosen according

to Shamblott et al. (2001). For the analysis of alkaline phosphatase

mRNA, speci®c oligonucleotides and PCR conditions were chosen

Human Reproduction Vol.18, No.7 pp. 1489±1493, 2003 DOI: 10.1093/humrep/deg279

ã European Society of Human Reproduction and Embryology 1489

at Serials Departm

ent on October 28, 2014

http://humrep.oxfordjournals.org/

Dow

nloaded from

according to Pittenger et al. (1999). As an internal standard, the

expression of 18S rRNA was analysed.

Western blot analyses

For Western blot analyses, protein extracts were prepared in buffer

containing 20 mmol/l HEPES pH 7.9, 0.4 mol/l NaCl, 2.5% glycerol,

1 mmol/l EDTA, 1 mmol/l phenylmethylsulfonyl ¯uoride, 0.5 mmol/l

NaF, 0.5 mmol/l Na3VO4, 0.02 mg/ml leupeptin, 0.02 mg/ml aprotinin,

0.003 mg/ml benzamidine chloride, 0.1 mg/ml trypsin inhibitor and

0.5 mmol/l dithiothreitol (DTT). Cells were lysed by freezing and

thawing, the extracts were centrifuged and the supernatants were

stored at ±70°C. Protein concentrations were determined using the

Bio-Rad protein assay reagent with bovine serum albumin as the

standard. Proteins were run on an SDS±polyacrylamide gel and

transferred to nitrocellulose. Blots were stained with Ponceau-S to

visualize equal amounts of loaded protein (for the method see also

Kubista et al., 2002). For immunodetection, the anti-Oct-4 antibody

C-10 (Santa Cruz, CA, USA) was used. As suggested by the

manufacturer of the antibody, the rodent teratocarcinoma cell line

F9 has been analysed as an Oct-4-positive control.

Immunocytochemistry

For immunocytochemical analyses of cellular Oct-4 expression, cells

were ®xed in cold methanol/acetone (1:1) and incubated with anti-

Oct-4 antibody C-10 (Santa Cruz) overnight at 4°C. Thereafter, cells

were washed, incubated with a biotinylated secondary antibody

(B7401, Sigma), washed again, and incubated with ExtrAvidin-Cy3

conjugate (red) (E4142, Sigma). For double staining, a second

incubation with the cyclin A-speci®c antibody H-432 (Santa Cruz)

was performed. This antibody was detected directly with a ¯uorescein

isothiocyanate (FITC)-conjugated anti-rabbit antibody (green)

(F9887, Sigma). Cell nuclei were counterstained with 4¢,6-

diamidino-2-phenylindole (DAPI; blue).

Results

RT±PCR was performed with 11 independent amniocentesis

samples. In ®ve samples, Oct-4 mRNA expression was

detectable. In Figure 1 representative samples are presented.

To obtain more information regarding the nature of these cells

and their potential, we investigated stem cell factor, vimentin

and alkaline phosphatase mRNA expression. All three of these

molecules are markers for pluripotent stem cells (for a detailed

description of such markers see http://www.nih.gov/news/

stemcell/scireport.htm). Stem cell factor (Martin et al., 1990) is

known to be expressed in ES cells, EC cells, haematopoietic

stem cells and mesenchymal stem cells. Vimentin is known to

be expressed in ectodermal, neural and pancreatic progenitor

cells (compare also Shamblott et al., 2001). Alkaline

phosphatase (compare Pittenger et al., 1999) expression can

be found in ES cells and in EC cells. By analysing six

independent amniotic ¯uid cell samples, we found that all the

three Oct-4-positive samples also expressed these three

markers, whereas the three Oct-4-negative cell samples did

not (Figure 1).

Western blot analyses demonstrated Oct-4 protein expres-

sion in samples with positive RT±PCR results (Figure 2).

Immunocytochemical analyses of positive samples revealed

that ~0.1±0.5% of amniotic ¯uid cells expressed Oct-4 protein

(Figures 3 and 4). The low number of Oct-4-expressing cells

can be assumed to be the explanation for the low expression

levels in the amniotic ¯uid samples detected by Western blot

analyses (Figure 2). In agreement with the known localization

of this POU transcription factor (Pesce and SchoÈler, 2001),

Oct-4 signals were detected in the nucleus (Figures 3 and 4).

To investigate further the question of whether the Oct-4-

positive amniotic ¯uid cells are actively dividing, we per-

formed double immunostaining experiments with anti-cyclin A

antibody. The cell cycle regulator cyclin A is only expressed in

proliferating cells (for a review see HengstschlaÈger et al.,

1999). Our ®nding that cyclin A protein is expressed in Oct-4-

positive amniotic ¯uid cells demonstrates that these cells

proliferate (Figure 4).

Discussion

A major aim in stem cell research is the identi®cation of new

sources of stem cells with as yet unknown properties. There is

no doubt as to the importance of Oct-4 for the maintenance of

stem cells. Each established mammalian pluripotent stem cell

line expresses Oct-4, which rapidly disappears when the cells

differentiate (Pesce and SchoÈler, 2001). ``If you wanted a

marker of pluripotent stem cells, you wouldn't go far wrong if

you picked the POU-domain transcription factor, Oct-4.''

(Donovan, 2001). Accordingly, we have analysed human

amniotic ¯uid cell samples for the expression of this marker.

For RT±PCR detection of Oct-4 mRNA, two independent sets

of oligonucleotides have been used, which previously were

shown to be speci®c for the human sequence of this transcrip-

tion factor (Monk and Holding, 2001; Xu et al., 2001). In the

Figure 1. Cells in human amniotic ¯uid express Oct-4 mRNA.RT±PCR analyses of Oct-4 mRNA expression in a representativesample of amniotic ¯uid cells containing Oct-4-positive cells (AFCA) and in a representative sample of Oct-4-negative amniotic ¯uidcells (AFC B). Two different sets of oligonucleotides have beenused, both known to be speci®c for human Oct-4 (Monk andHolding, 2001; Xu et al., 2001). As an internal standard, theexpression of 18S rRNA (18S) was analysed. In addition, in thesesamples, expression of stem cell factor (SCF), vimentin and alkalinephosphatase (ALP) mRNA has been investigated via RT±PCR.

A.-R.Prusa et al.

1490

at Serials Departm

ent on October 28, 2014

http://humrep.oxfordjournals.org/

Dow

nloaded from

RT±PCR-positive samples, we detected the Oct-4 protein of

the correct size using a speci®c antibody in Western blot

analyses. To check further whether the amniotic ¯uid cells

exhibit the correct intracellular localization of this transcription

factor, we performed immunocytochemical investigations. The

observed nuclear localization again con®rms that the expres-

sion of the correct molecule has been analysed in our study. In

addition, this ®nding furthers suggest that Oct-4 expressed in

amniotic ¯uid cells is functional.

The immunocytochemical analyses were also of importance

in answering the question of exactly how many cells within one

positive amniotic ¯uid cell sample express Oct-4. It has been

reported previously that some adult human tissues express low

levels of Oct-4 as analysed by RT±PCR. Whether Oct-4 protein

expression is detectable in these tissues remained elusive

(Takeda et al., 1992). In Western blot analyses, we found that

speci®c human tissues samples, others than amniotic ¯uid

cells, were positive for Oct-4 expression in RT±PCR but did

not express Oct-4 protein (data not shown). This ®nding

provides evidence for post-transcriptional control of Oct-4

expression in speci®c human cells. In the study presented here,

we observed that a subpopulation within amniotic ¯uid cell

samples can be found to be Oct-4 positive. The fact that only

~0.1±0.5% of the cells within such a positive sample express

this transcription factor, together with the ®nding that only

about half of the analysed amniotic ¯uid samples contained

Oct-4-positive cells, suggests that (i) the Oct-4-positive signals

are speci®c for a distinct subpopulation within an amniotic

¯uid cell sample and are not due to a low background of Oct-4

expression in all different cell types existing in such a sample;

and (ii) Oct-4-positive cells cannot be detected in every

analysed sample. The reasons for this observation still need to

be clari®ed.

When ES cells are triggered to differentiate, Oct-4 is down-

regulated. Loss of Oct-4 expression at the blastocyst stage

causes the cells of the inner cell mass to differentiate into

trophoectoderm cells (Pesce and SchoÈler, 2001). Experiments

in mouse models indicate that an Oct-4 expression level of

~50±150% of the endogenous amount in ES cells is permissive

for self-renewal and maintenance of totipotency. Up-regulation

above these levels causes stem cells to express genes involved

in the lineage differentiation of primitive endoderm (Niwa

et al., 2000). At present, this model has not been proven in

humans. However, it could be considered of interest to

compare the level of expression of Oct-4-positive amniotic

¯uid cells with that in different stem cell types and lines to

obtain further insights into what type of stem cells are detected

in amniotic ¯uid cell samples. Experiments with human ES

cells or stem cell lines are forbidden in Austria.

To obtain further information about amniotic ¯uid cells in

connection with stem cell research, we sought to answer two

important questions. (i) Can other markers, known to be

expressed in pluripotent human stem cells, be found in

amniotic ¯uid samples? (ii) Are the Oct-4-positive amniotic

¯uid cells actively dividing? RT±PCR analyses revealed that in

amniotic ¯uid cell samples containing Oct-4 positive cells, the

Figure 3. Nuclear localization of Oct-4 protein in amniotic ¯uidcells. Immunocytochemical analyses revealed that a subpopulationof Oct-4 cells can be found in human amniotic ¯uid cell samples. Inthe ®gure, one Oct-4-positive cell and one Oct-4-negative cell canbe seen. The Oct-4-speci®c signal in positive cells was located inthe nucleus. A phase contrast photomicroscopy analysis (upperpicture) and a ¯uorescence microscopy analysis of the Oct-4speci®c signal (lower picture) are shown.

Figure 2. Cells in human amniotic ¯uid express Oct-4 protein.Western blot analysis of Oct-4 protein expression in a representativesample of amniotic ¯uid cells containing Oct-4-positive cells (AFC1) and a representative sample of Oct-4-negative amniotic ¯uid cells(AFC 2). As a positive control, the rodent teratocarcinoma cell lineF9 has been analysed.

Oct-4-positive amniotic ¯uid cells as a source of stem cells?

1491

at Serials Departm

ent on October 28, 2014

http://humrep.oxfordjournals.org/

Dow

nloaded from

mRNAs of three genes, known to be expressed in pluripotent

human stem cells, can be detected: stem cell factor, vimentin

and alkaline phosphatase. Double staining experiments further

demonstrated that Oct-4-positive amniotic ¯uid cells can

actively divide, which was proven by the detection of cyclin

A expression. However, ®nal proof for a putative pluripotency

of such Oct-4-positive amniotic ¯uid cells would include

experiments to isolate them and to differentiate them into

different lineages. Such experiments are planned in our

laboratory. We believe this new source of human stem cells

to be interesting, even if these cells could be proven to harbour

the potency to differentiate only in a speci®c subset of lineages.

In this respect, it will also be of interest to try to elucidate the

origin of the Oct-4-positive cells in the human amniotic ¯uid.

There exist a lot of different possibilities, including, for

example, aberrantly migrating primordial germ cells.

Another important marker for human pluripotent stem cells

is telomerase activity (for a detailed description of such

markers see http://www.nih.gov/news/stemcell/scireport.htm).

All ES cell lines express high levels of telomerase, the enzyme

that helps to maintain telomeres which protect the ends of

chromosomes. Telomerase activity and long telomeres are

characteristic of proliferating cells in embryonic tissues and of

germ cells. Human non-transformed somatic cells, however, do

not show telomerase activity and their telomeres are consid-

erably shorter (Amit et al., 2000). Interestingly, telomerase

activity can be detected in human amniotic ¯uid cell samples

(Mosquera et al., 1999). This observation is in agreement with

the assumption that such samples can contain stem cells of high

potency, as suggested by our data.

We believe the ®ndings reported here, together with the

recent demonstration of the successful usage of amniotic ¯uid

cells for tissue engineering approaches (Kaviani et al., 2001),

to be encouraging for the further investigation of human

amniotic ¯uid as a putative new source of stem cells with high

potency.

References

Amit, M., Carpenter, M.K., Inokuma, M.S., Chiu, C.P., Harris, C.P., Waknitz,M.A., Itskovitz-Eldor, J. and Thompson, J.A. (2000) Clonally derivedhuman embryonic stem cell lines maintain pluripotency and proliferativepotential for prolonged periods of culture. Dev. Biol., 227, 271±278.

Figure 4. Oct-4-positive amniotic ¯uid cells express cyclin A. Amniotic ¯uid cell nuclei were stained with DAPI (upper panel). These cellswere analysed for Oct-4 protein expression (middle panel) and for the expression of cyclin A protein (lower panel).

A.-R.Prusa et al.

1492

at Serials Departm

ent on October 28, 2014

http://humrep.oxfordjournals.org/

Dow

nloaded from

Donovan, P.J. (2001) High Oct-ane fuel powers the stem cell. Nature Genet.,29, 246±247.

Gosden, C.M. (1983) Amniotic ¯uid cell types and culture. Br. Med. Bull., 39,348±354.

HengstschlaÈger, M., Braun, K., Soucek, T., Miloloza, A. and HengstschlaÈger-Ottnad, E. (1999) Cyclin-dependent kinases at the G1±S transition of themammalian cell cycle. Mutat. Res., 436, 1±9.

Hoehn, H. and Salk, D. (1982) Morphological and biochemical heterogeneityof amniotic ¯uid cells in culture. Methods Cell Biol., 26, 11±34.

Kaviani, A., Perry, T.E., Dzakovic, A., Jennings, R.W., Ziegler, M.M. andFauza, D.O. (2001) The amniotic ¯uid as a source of cells for fetal tissueengineering. J. Pediatr. Surg., 36, 1662±1665.

Kubista, M., Rosner, M., Kubista, E., Bernaschek, G. and HengstschlaÈger, M.(2002) Brca1 regulates in vitro differentiation of mammary epithelial cells.Oncogene, 21, 4747±4756.

Martin, F.H., Suggs, S.V., Langley, K.E., Lu, H.S., Ting, J., Okino, K.H.,Morris, C.F., McNiece, I.K., Jacobsen, F.W., Mendiaz, E.A., et al. (1990)Primary structure and functional expression of rat and human stem cellfactor DNAs. Cell, 63, 203±211.

Milunsky, A. (1979) Amniotic ¯uid cell culture. In Milunsky, A. (ed.) GeneticDisorder and the Fetus. Plenum Press, New York, pp. 75±84.

Monk, M. and Holding, C. (2001) Human embryonic genes re-expressed incancer cells. Oncogene, 20, 8085±8091.

Mosquera, A., Fernandez, J.L., Campos, A., Goyanes, V.J., Ramiro-Diaz, J.R.and Gosalvez, J. (1999) Simultaneous decrease of telomere length andtelomerase activity with ageing of human amniotic ¯uid cells. J. Med.Genet., 36, 494±496.

Niwa, H., Miyazaki, J-i. and Smith, A.G. (2000) Quantitative expression ofOct-3/4 de®nes differentiation, dedifferentiation or self renewal of ES cells.Nature Genet., 24, 372±376.

Pesce, M. and SchoÈler, H.R. (2001) Oct-4: gatekeeper in the beginnings ofmammalian development. Stem Cells, 19, 271±278.

Pittenger, M.F., Mackay, A.M., Beck, S.C., Jaiswal, R.K., Douglas, R., Mosca,J.D., Moorman, M.A., Simonetti, D.W., Craig, S. and Marshak, D.R. (1999)Multilineage potential of adult human mesenchymal stem cells. Science,284, 143±147.

Prusa, A.R. and HengstschlaÈger, M. (2002) Amniotic ¯uid cells and humanstem cell research: a new connection. Med. Sci. Monitor, 8, 253±257.

Shamblott, M.J., Axelman, J., Little®eld, J.W., Blumenthal, P.D., Huggins,G.R., Cui, Y., Cheng, L. and Gearhart, J.D. (2001) Human embryonic germcell derivatives express a broad range of developmentally distinct markersand proliferate extensively in vitro. Proc. Natl Acad. Sci. USA, 98, 113±118.

Takeda, J., Seino, S. and Bell, G.I. (1992) Human Oct3 gene family: cDNAsequences, alternative splicing, gene organization, chromosomal location,and expression at low levels in adult tissues. Nucleic Acids Res., 20,4613±4620.

Xu, C., Inokuma, M.S., Denham, J., Golds, K., Kundu, P., Gold, J.D. andCarpenter, M.K. (2001) Feeder-free growth of undifferentiated humanembryonic cells. Nature Biotechnol., 19, 971±974.

Submitted on October 2, 2002; resubmitted on February 13, 2003; accepted onMarch 18, 2003

Oct-4-positive amniotic ¯uid cells as a source of stem cells?

1493

at Serials Departm

ent on October 28, 2014

http://humrep.oxfordjournals.org/

Dow

nloaded from