ISSCR :: Scientists : Topic of the Month March 2004

When injected into mice, they formed teratomas containing differentiation into all three germ layers. Munsie et al. demonstrated extensive contribution by such cells to numerous organs in chimeric fetuses and pups.

Subsequently, Wakayama et al. demonstrated that ntES cell lines they derived could also contribute to the germline when injected into blastocysts. Moreover, they were able to use the ntES cell nuclei to successfully clone live mice, demonstrating full pluripotency of their ntES cells.

After the first successful derivation of human embryonic stem cell lines, it was just a matter of time and technical improvements, before cloned ES cell lines would be derived from human blastocysts.

In their report, Hwang et al. derived ntES cells by performing autologous somatic cell nuclear transfer. The group compared several conditions and report cloned human blastocysts from 19-29% of reconstructed eggs, comparable to rates in cattle and pigs. However, from 20 ICMs they were only able to derive one ES cell line, compared to previously reported 17-50% of extracted human ICMs, but the authors acknowledge that some of the blastocysts may have been aneuploid, which could contribute to their low cell line derivation rate.

The ntES cells maintained undifferentiated morphology for >70 passages, displayed bi-parental and not unimaternal expression of certain imprinted genes and maintained a normal karyotype. The latter finding suggests that this cell line had escaped the molecular problems often associated with the nuclear transfer process (see below). One reason for successful derivation of a karyotypically normal cell line could come from the altered enucleation protocol, as Hwang et al. squeezed the oocytes to extrude the nuclei through a small hole in the zona pellucida, shortly after the appearance of the first polar body, rather than aspirating the nuclei with a glass pipette.

Before heading into the clinic, it will become important to clearly address molecular problems that have been reported after nuclear transfer in animal models. Several reports on molecular characterization of cells developing after nuclear transfer have made it clear that gene expression patterns as well as cellular structures appear to be disrupted by the procedure.

In their review of the literature on bovine nuclear transfer, Niemann et al. summarized defective gene expression patterns for numerous genes after nuclear transfer.

When analyzing gene expression of Oct4 and Oct10-related genes in murine cumulus cell-derived cloned blastocysts by microarray, Bortvin et al. found that a worrying 38% of analyzed genes were incorrectly expressed. This was not the case in cloned blastocysts derived from ES cell nuclei or in normal control embryos.

Simerly et al. reported disarrayed mitotic spindles with misaligned chromosomes in primate cells dividing after nuclear transfer, resulting in aneuploid embryos. The group went on to show that spindle removal depletes the ooplasm of two proteins crucial for mitotic spindle pole formation.

Conversely, two reports indicate normal development when looking at one specific aspect of chromatin structure and the reprogramming of the donor genome, in the mouse. The two groups investigated the replacement pattern of oocyte-specific H1 linker histone H1F00. During normal early development the oocyte-specific H1F00 is progressively replaced by somatic H1 in the chromatin.

After nuclear transfer, Gao et al. and Teranishi et al. report a rapid (within minutes) replacement of somatic H1 on the DNA of the incoming nucleus by the oocytes-specific H1F00. This is then followed by the normal progressive loss of H1F00 and assembly of somatic H1 into the chromatin of the transferred nucleus. These results suggest that leaving time between transfer of the nucleus and activation of the oocyte may increase success by allowing these events to happen. Accordingly, it had been reported previously that the time lapse between introduction of the nucleus and activation of the oocyte matters for following development.

Now that nuclear transfer has been achieved successfully in human cells, the next question becomes: where and how to procure the large amount of oocytes that will be required? This may particularly become a problem for research, as more efficient methods for derivation of ntES cell lines will have to be developed.

Several avenues are being pursued. One option is parthenogenesis. This approach could multiply the number of available eggs by two and could be a useful tool for research, as it would provide duplicate samples for meaningful experimentation. One group described the successful derivation of a parthenogenetic ES cell line from Macaca fascicularis.

Two groups reported interspecies nuclear transfer, transferring human DNA into enucleated non-human oocytes. Chen et al. described generation of ntES cells after transfer of human nuclei into rabbit oocytes. The obtained cells are described as human as assessed by karyotype, in situ hybridization, PCR and immunocytochemistry with specific human probes. Chang et al. reported generation of blastocysts after transfer of human nuclei from umbilical cord fibroblast into enucleated bovine oocytes with only the bovine mitochondrial DNA persisting beyond the morula stage.

Yet another source of eggs could come from the generation of oocytes from ES cells, as described by Hubner et al. in the mouse. The in vitro generated oocytes ranged within the size of natural oocytes, had a thin zona pellucida, re-expressed Oct-4, expressed mRNA for the oocyte-specific markers ZP2 and 3, but failed to express ZP1, possibly accounting for the fragile zona pellucida observed on these oocytes. Upon prolonged culture, blastocyst-like structures formed, presumably parthenogenotes, suggesting that culture conditions induced cleavage. Oocytes and blastocyst-like structures could be derived from female and male ES cells. The group will now assess whether those cells could be used as starting material for derivation of ntES cell lines.

But the very recent discovery of germline stem cells in the postnatal ovaries of mice now raises the hope that there could be germline stem cells in the human ovaries as well. Isolation of such cells and their characterization may be a first step towards bulk in vitro oocytes production.

In an elegant study, Rideout et al. demonstrated the feasibility of treatment of a genetic disorder by combining therapeutic cloning and gene therapy, both goals for future applications in humans. The group derived ntES cells from Rag2¯/¯ mice by transfer of somatic tail-tip nuclei of this mouse into enucleated murine embryos, repaired one defective allele of the Rag2 gene by homologous recombination, derived hematopoietic progenitors from these Rag2+R/¯ ntES cells in vitro (after retroviral transduction of HoxB4), and transplanted the cells into the Rag2¯/¯ mice.

While live cloned mice generated from the genetically repaired ntES cells had fully reconstituted lymphoid compartments, reconstitution was not possible in Rag2¯/¯ mice transplanted with in vitro derived hematopoietic progenitors due to low MHC expression on the differentiated ntES cells and consequent destruction by host NK cells. Depletion of NK cells prior to transplantation partially overcame this, but transplantation into Rag2¯/¯ mice on a gC chain-deficient background (devoid of NK cells) resulted in reconstitution, albeit predominantly myeloid. The authors suggest that the presence of the HoxB4 gene may have skewed development towards myeloid lineage, as had been reported previously.

Barberi et al. reported dopaminergic neurons derived in vitro from ntES cell, and were able to correct a mouse model of Parkinson's disease by injecting these neurons into the ipsilateral striatum of the mice. Two months after transplantation, the grafts had extended over a large portion of the host striatum, with alleviation of behavioral deficits.

Neural subtype specification of fertilization and nuclear transfer embryonic stem cells and application in parkinsonian mice. ABSTRACT
Barberi T, Klivenyi P, Calingasan NY, Lee H, Kawamata H, Loonam K, Perrier AL, Bruses J, Rubio ME, Topf N, Tabar V, Harrison NL, Beal MF, Moore MA, Studer L (2003). Nat Biotechnol 21:1200.

These reports beg the question of potential for degeneration of the transplanted cells. While teratoma formation has been well documented upon injection of undifferentiated ES cells, less has been reported on teratoma formation by cells differentiated from ES cells. While Barberi et al. found no signs of aberrant differentiation or teratoma formation, Sipione et al. did. The group differentiated insulin-expressing cells in vitro from mouse ES cells and injected the cells into streptozotocin-treated SCID mice (diabetes model). In their hands, blood glucose levels were not reduced, and upon sacrifice the mice exhibited what resembled teratomas.

Insulin expressing cells from differentiated embryonic stem cells are not beta cells. ABSTRACT
Sipione S, Eshpeter A, Lyon JG, Korbutt GS, Bleackley RC (2004). Diabetologia Feb 14 [Epub ahead of print]

Human ES cells in the Clinic?

Despite the forward leap in human embryonic stem cell research supplied by Hwang et al., much remains to be done. Before heading into the clinic, cell therapy with human ES cells will have to be evaluated carefully. Risk of in vivo teratoma formation will have to be evaluated, as well as the homing capacities of the injected cells. The latter may be dramatically different from what was observed in animal models injected with human cells. When injecting human cells into the human environment receptors, chemokines, cytokines, growth factors and adhesion molecules will be able to interact in a cognate and much more specific manner. This may result in improved homing, but may also result in new cues for the injected cells, potentially resulting in altered phenotype.

An interdisciplinary group of experts in science, law and philosophy deliberated on safety and ethical issues of using human ES cells in clinical settings. They reported their conclusions in the review below. Important points made by the group include the risk of infectious disease, transfer of genetic disorders and quality control of stem cell lines and their derivatives. Moreover, the group points out the potential for "misdifferentiation" of the cells in vivo, as well as side effects triggered by potential migration of the cells to other tissues than the target tissues, and potential tumor formation. The importance of immune rejection, as well as approaches to the design of clinical trials are also discussed in detail.

Safety issues in cell-based intervention trials. ABSTRACT
Dawson L, Bateman-House AS, Mueller Agnew D, Bok H, Brock DW, Chakravarti A, Greene M, King PA, O'Brien SJ, Sachs DH, Schill KE, Siegel A, Solter D, Suter SM, Verfaillie CM, Walters LB, Gearhart JD, Faden RR (2003). Fertil Steril 80:1077.