Editorial Summary

Making embryonic stem cells by moving a differentiated nucleus into an oocyte is now possible in monkeys. Credit: Shoukhrat M. Mitalipov

Shoukhrat Mitalipov and colleagues show that the nuclei of adult monkey cells can be placed in a monkey oocyte, grown into an embryo, and used to make embryonic stem cells. Though the same procedure was demonstrated in mice about seven years ago, many failed attempts led to speculation that differentiated primate cells could not perform a similar feat. The success in monkeys suggests that a similar accomplishment is possible in humans, raising the likelihood of creating patient-specific stem cells.

The Oregon group generated two embryonic stem cell lines from 304 oocytes taken from 14 rhesus monkeys, a fairly low 0.7% success rate. However, the factors behind this success are still not clear. Should most of the credit go to a modified technique or a very skilled scientist? And what factors might lead to a higher efficiency? Here, we reveal what a panel of four anonymous peer reviewers had to say about this breakthrough.

Read below to see a panel of experts' comments (in black), and responses from the authors (in italics). Table and reference numbers refer to those in the research article.

Care to comment? Blog here on The Niche

The Experts' Corner: Experts in embryonic stem cells, nuclear transfer and genotyping weigh in

Nuclear Transfer Expert

This manuscript describes the first derivation of ES cells from cloned primate embryos. The authors are to be congratulated for their achievement. The results are of great interest to those concerned with the use of nuclear transfer in stem cell research. In view of the recent history of publications in this area of research it is essential that the observations are confirmed by independent tests and subject only to that confirmation the paper should be published rapidly.

At some point in the paper it is essential to make the point that further research is required to define the critical factors in this success. Nothing presented in this manuscript assesses the role of the novel approach to enucleation.

Author's response

We agree with the reviewer and have added the following statement to the discussion:

“We speculate that the impaired blastocyst formation rate following conventional SCNT in primates may result from one or more of the following factors: Hoechst 33342 and/or UV damage to the relatively transparent primate oocyte, Hoechst 33342/UV induced oocyte activation and/or MPF degradation, reaction of the residual Hoechst 33342 in cytoplasts with the introduced donor cell DNA thereby impairing reprogramming and/or Hoechst 33342 contact with mitochondrial DNA, thus reducing cytoplast mitochondrial function. Further research is required to define the critical factors and mechanisms determining developmental success following different spindle removal and SCNT methodologies.”

Nuclear Transfer and ES cell expert

To my knowledge, this is the first report of the generation of a SCNT ES line in monkeys. SCNT in primates has hitherto either been shrouded in fraud or totally unsuccessful. Simerly et al (ref 11 in manuscript) concluded that reconstructed, monkey SCNT blastocysts were karyotypically abnormal and speculated that this was a consequence of faulty spindle assembly following nuclear transfer. They further speculated that human SCNT would be similarly unsuccessful. The claims of the present manuscript would render these previous “doom and gloom” claims premature. Moreover, a robust method for SCNT-mediated ES formation would be encouraging for those considering the use of similar methods for human SCNT.

The authors used a “modified” SCNT procedure to effect a very inefficient generation of monkey SCNT ES cells. 304 oocytes were used to make 20 morphologically low grade blastocysts. From these, 2 lines were made, one of which was karyotypically abnormal. The overall efficiency was 0.7%, much lower than that typically obtained using SCNT in mouse. Validation of the provenance of the two lines involved a variety of techniques and was thorough. Microsatellite and SNP analysis demonstrated the origin of the ES lines from the donor fibroblasts; mitochondrial DNA analysis confirmed the recipient oocyte origin of the mitochondria. The pluripotency of the lines was indicated by standard teratoma formation assays as well as limited directed differentiation assays for the cardiac and neuronal lineages. I thought the general histology and immunostaining performed was rather poor in quality although this may have been an electronic data problem. Transcriptional profiling was also performed alongside a range of sensible controls including donor fibroblasts and IVF-generated ES lines. These studies generally confirmed the ES like nature of the SCNT ES lines. I would have like to have seen some karyotype data from blastocysts made by this method. Perhaps this is in the work in press; one has to assume that their new method has greatly improved the karyotypic stability. If they just got lucky with the one normal line they claimed, this might make independent corroboration of these data problematic. However, despite these reservations regarding the histology and reproducibility, I was convinced by the totality of the data that these authors had achieved their claimed objective-the generation of a monkey SCNT line

Author's response

We have now included direct references to our paper recently published in Human Reproduction and have added the following paragraphs to the discussion:

“We speculate that the impaired blastocyst formation rate following conventional SCNT in primates may result from one or more of the following factors: Hoechst 33342 and/or UV damage to the relatively transparent primate oocyte, Hoechst 33342/UV induced oocyte activation and/or MPF degradation, reaction of the residual Hoechst 33342 in cytoplasts with the introduced donor cell DNA thereby impairing reprogramming and/or Hoechst 33342 contact with mitochondrial DNA, thus reducing cytoplast mitochondrial function. Further research is required to define the critical factors and mechanisms determining developmental success following different spindle removal and SCNT methodologies.

Recognising the importance of high quality cytoplasts for the successful reprogramming, we have been seeking non-invasive approaches for spindle detection and removal. “Blind” enucleation techniques involving “squishing” 20 or “one step manipulation” (OSM) 21 are intrinsically faulty, at least in our hands, because they fail to enucleate all oocytes. In fact, our initial effort to derive ntESCs from blastocysts produced with the OSM protocol resulted in the isolation of parthenogenetic ESCs, almost certainly due to failed spindle removal. Fortunately, recent developments in high-performance imaging resulted in an Oosight™ spindle imaging system supporting non-invasive, rapid and highly efficient real time enucleation of primate oocytes.”

Reviewers' response

In the discussion section, the authors write: “We observed a 16% blastocyst formation rate, significantly higher than rates obtained with conventional SCNT protocols (1%)10, but somewhat lower than previous results reported in a comparative study using a one-step approach14. In this latter case, a single, and different, fibroblast cell line was employed as the nuclear donor source and spindle removal was accomplished under direct visualization using Differential Interference Contrast (DIC) optics. Differences in the SCNT blastocyst rates using these two protocols may be donor cell dependent (Mitalipov et al., in press) or may reflect a higher parthenote production rate under the direct visualization protocol.”

I was unclear about the mention of parthenotes in the region I have highlighted. Are the authors implying that the one-step approach paper did not rule out the accidental formation of parthenotes?

In our hands the one-step SCNT approach has only resulted in the production of parthenote-derived ESCs (all three ESC lines derived with this technique were parthenote derived). We include the following sentence to clarify our observations:

““Blind” enucleation techniques involving “squishing” 20 or “one step manipulation” (OSM) 21 are intrinsically faulty, at least in our hands, because they fail to enucleate all oocytes. In fact, our initial effort to derive ntESCs from blastocysts produced with the OSM protocol resulted in the isolation of parthenogenetic ESCs, almost certainly due to failed spindle removal. Fortunately, recent developments in high-performance imaging resulted in an Oosight™ spindle imaging system supporting non-invasive, rapid and highly efficient real time enucleation of primate oocytes.

With reference to the potential immunogenicity of the CERS ESC line to the donor, on account of the foreign mitochondrial content, the authors write:

“It should be noted that further research is required to investigate whether any immune response might occur in the donor animal resulting from the epitopes derived from the allogenic recipient oocyte mitochondrial contribution; although preliminary research with cloned bovine cells and tissue with allogeneic mtDNA suggests grafting back into the nuclear donor organism can occur without destruction by the immune system24”

I would point out that the cloned porcine skin transplantation work of John Logan and his colleagues (Martin et al 2003 Cloning Stem Cells 5 110-121) is a better example for showing that the mitochondrial contribution is irrelevant

We have included the suggested Martin et al 2003 reference in the relevant section.

In Supplementary Fig 3, I have to take on trust the authors designation of the SCNT blastocysts as “low grade”. The high grade monkey embryos look different but are nowhere as good as Class 1 or 2 human blastocysts. This could be a species thing. I do not have that degree of knowledge.

Human blastocysts have a significantly different morphology to rhesus monkey blastocysts, the human trophectodermal cell borders appear more clearly defined and the rhesus monkey embryonic cytoplasm appears darker and more granular than human embryonic cytoplasm. It could be species-specific differences such as these that the reviewer is referring to when discussing embryo grading.

In conclusion, data wise I believe that this paper reports for the first time a bone fide SCNT-generated macaque ES line. As such it is probably suitable for publication in Nature. My reservation here concerns the very low efficiency of the technique which will make it very difficult for other labs to reproduce. 304 oocytes were used; this number represents a large use of stimulated monkeys, and requires huge resources. It is therefore not clear whether it was the “refined” method that contributed to the success or whether they just got lucky through statistical chance. I would be reassured if the authors had shown in this (or alternatively in their paper “in press” that they were getting largely karyotypically normal blastocysts. It is important to know the answer to this.

We agree with the reviewer that karyotyping individual SCNT blastocysts would be an interesting subject for future research and so we have added the following sentence to the discussion:

“Interesting areas for future research include karyotyping, methylation and expression analysis of individual SCNT blastocysts.”

I believe the paper should be rewritten so as to remove some of the more obvious and repetitive allusion to the human SCNT situation. Authors should be specific as to the refinements they have developed along with some mention of why they think these work and why previous attempts had failed.

We have shortened this section of the discussion in an effort to comply with the reviewers suggestion. However due to the relevance of this primate SCNT research to human SCNT we would like to keep some sections that refer to human SCNT intact. We agree with the reviewer regarding the suggestion that we should include a section explaining why we think the Oosight enucleation improves rate of primate SCNT blastocyst formation. We have added the following statements to the discussion:

“We speculate that the impaired blastocyst formation rate following conventional SCNT in primates may result from one or more of the following factors: Hoechst 33342 and/or UV damage to the relatively transparent primate oocyte, Hoechst 33342/UV induced oocyte activation and/or MPF degradation, reaction of the residual Hoechst 33342 in cytoplasts with the introduced donor cell DNA thereby impairing reprogramming and/or Hoechst 33342 contact with mitochondrial DNA, thus reducing cytoplast mitochondrial function. Further research is required to define the critical factors and mechanisms determining developmental success following different spindle removal and SCNT methodologies.

Recognising the importance of high quality cytoplasts for the successful reprogramming, we have been seeking non-invasive approaches for spindle detection and removal. “Blind” enucleation techniques involving “squishing” 20 or “one step manipulation” (OSM) 21 are intrinsically faulty, at least in our hands, because they fail to enucleate all oocytes. In fact, our initial effort to derive ntESCs from blastocysts produced with the OSM protocol resulted in the isolation of parthenogenetic ESCs, almost certainly due to failed spindle removal. Fortunately, recent developments in high-performance imaging resulted in an Oosight™ spindle imaging system supporting non-invasive, rapid and highly efficient real time enucleation of primate oocytes.”

Human ES cell expert

The paper reports somatic cell nuclear transfer (SCNT) in the rhesus macaque, using skin fibroblasts of an adult male electrofused into enucleated oocytes of the same species. Two rhesus macque embryonic stem cell (CRES) lines were developed from 304 SCNT procedures (was this 304 oocytes?) using differential interference contrast (DIC) optics (rather than the previous method that involved Hoechst staining and UV light to identify the metaphase II spindle). The implied assumption was that DIC optics was the factor that significantly improved the 1% blastocyst rate to 16% (Suppl. Table 1). However, the authors state that the improvement may be due to the donor cells (Discussion).

1. The evidence that the improvement is donor cell dependent, not due to DIC optics or SCNT expertise, should be provided because this is critical and central to the study. This is quoted to be Mitalipov et al. in press. It is not appropriate to quote unpublished data in support of critical interpretations, experimental evidence or necessary methods. This paper has many references to 'in press or review' articles, which should be either deleted or properly referenced as accepted or published articles. While they are present in the paper, it is impossible to judge the validity or otherwise of the claims, interpretations or methods.

We have now included direct references to our paper published in Human Reproduction and have also added the following paragraphs to the discussion:

“We speculate that the impaired blastocyst formation rate following conventional SCNT in primates may result from one or more of the following factors: Hoechst 33342 and/or UV damage to the relatively transparent primate oocyte, Hoechst 33342/UV induced oocyte activation and/or MPF degradation, reaction of the residual Hoechst 33342 in cytoplasts with the introduced donor cell DNA thereby impairing reprogramming and/or Hoechst 33342 contact with mitochondrial DNA, thus reducing cytoplast mitochondrial function. Further research is required to define the critical factors and mechanisms determining developmental success following different spindle removal and SCNT methodologies.

Recognising the importance of high quality cytoplasts for the successful reprogramming, we have been seeking non-invasive approaches for spindle detection and removal. “Blind” enucleation techniques involving “squishing” 20 or “one step manipulation” (OSM) 21 are intrinsically faulty, at least in our hands, because they fail to enucleate all oocytes. In fact, our initial effort to derive ntESCs from blastocysts produced with the OSM protocol resulted in the isolation of parthenogenetic ESCs, almost certainly due to failed spindle removal. Fortunately, recent developments in high-performance imaging resulted in an Oosight™ spindle imaging system supporting non-invasive, rapid and highly efficient real time enucleation of primate oocytes.

I don't think that the reference to, and comparisons with, parthenogenetic rhesus macque ESCs (RPES-4) is very helpful or useful. It is very likely that parthenogenetic ESCs are homozygous in the pericentromeric region of the chromosome, but heterozygous when distal to the centromere. This has been described in the mouse by G. Daley's group at Harvard. One would expect the SNPs to vary in parthenogenetic ESCs depending on their position in the chromosome. There was no variation in SNPs for the parthenogenetic ESCs. I suggest this data be deleted from the paper or further SNP analyses be undertaken to demonstrate they are genuinely parthenogenetic.

We agree with the reviewer and have replaced the parthenogenetic rhesus monkey ESC line (RPES-4) with an IVF-derived rhesus ESC line (ORMES-22) which we feel is a more useful control to demonstrate allele inheritance in the STR analysis:

“The genomic constitution of an IVF-derived rhesus monkey ESC line (ORMES-2217) and the ORMES-22 oocyte donor female (Female #3) and sperm donor male (Male #2) were also included to demonstrate STR allele inheritance (Table 1).”

In cattle SCNT, heteroplasmy of donor cell mitochondria may persist in blastocysts and even offspring (eg. Steinborn et al. Nature Genet. 2000, 25: 255–257) and in interspecies SCNT when monkey donor cells are introduced into rabbit oocytes (Yang et al., Mol Rep Dev 65, 396–401; Reproduction 127, 201–205). While it may be correct that all donor cell mitochondria are lost in the CRES lines, further analysis is deserved to confirm this, given the possibility of a minor mitochondrial contribution from the donor cells. It may be that only the more abundant mitochondrial DNA is amplified in the methods used - is this possible? The criticism could be that the finding of only oocyte mitochondrial DNA variants is very convenient for the demonstration of cytoplasmic origins of the CRES lines - as well it may be.

Any donor cell mitochondrial inheritance was too small to be observed as a smaller peak on the chromatograms following the sequencing reaction. This certainly does not completely rule out a minor donor cell mitochondrial heteroplasmy and we agree with the reviewer that analysis for heteroplasmy would be an interesting field for future research on these CRES lines. However, for the purposes of this study, our focus was on demonstrating oocyte mitochondrial inheritance to demonstrate that the CRES lines were in fact derived though SCNT and not accidentally produced parthenotes.

The mixed aneuploidy for loss of the Y chromosome and the presence of the isochromosome with 2 copies of the long arm of the Y chromosome in CRES-1 is interesting. It is possible that the single donor cell used had this chromosomal aneuploidy. Was there a detailed analysis of the donor fibroblast cell line to determine if there were any cells with similar or related Y chromosomal errors and aneuploidies?

* Cytogenetic analysis was performed on 20 metaphase cells from the donor fibroblasts from male #1, following standard GTW-banding procedures and we also performed fluorescence in situ hybridization (FISH) on metaphase cells from the donor somatic cell line utilizing BAC CH250-283K14, specific for the rhesus macaque Yq11.21 region. The results of both the G banding and FISH analysis demonstrate a normal male rhesus macaque chromosome complement in all donor cells analyzed (42, XY). We did not observe any donor cells with Y chromosomal errors or aneuploidies.

Genotyping Expert

Physicians and scientists looking for new treatments for diseases that are likely to benefit from stem cell therapy; ESC biologists looking for modifications to current techniques that are effective in primate cells; and possibly geneticists studying imprinting of ntESCs.

The authors state they have used nuclear reprogramming to create pluripotent ESCs from rhesus monkey SCNT embryos. The ESCs have morphology similar to IVF produced embryos, they express ESC markers, they are transcriptionally indistinguishable from control IVF produced ESCs and they possess the potential to differentiate into multiple cell types. Reprogramming primate somatic cells into ESCs is a significant step towards therapeutic cloning in primates.

It is significant in regards to the success of this methodology using primate somatic cells. The majority of papers published in this field use non-primate cells, especially mouse and bovine. This study uses rhesus monkey cells, which provde a model more similar to human ESCs. Use of somatic cells from rhesus monkey to reprogram ESCs appears to be novel.

The genetic data presented in this manuscript demonstrate that the cloned rhesus embryonic stem cell lines carry nuclear material from the male somatic nuclei donor and the cytoplasmic genetic material (mitochondrial DNA) from the oocyte donor. Generation in vitro of cardiomyocytes and neuronal cells strongly supports the authors' claims that the ESCs are pluripotent.

However, the authors claim that the CRES ESC cell lines are indistinguishable from IVF-produced ESCs. Work by Smith et al., PNAS 102: 17852, 2005 with bovine ESCs compared gene expression profiles of cloned blastocysts with the somatic donor cells and found that the cloned blastocysts resembled naturally fertilized embryos with <1% of genes showing differential expression and considered this degree of difference as “indistinguishable". Grouping CRES1 biological replicates, grouping ORMES10 replicates, and performing pairwise comparison of array shows more than 6% difference (1607 genes of 24,997) in gene expression. Is this amount of difference really considered “indistinguishable”?

When we compared control ORMES-10 and control ORMES-22 biological replicates we observed a greater transcriptional variation than was observed between ORMES-10 and the CRES-1 and CRES-2 biological replicates. Thus the control intra-group variation (control vs control) was greater than the test inter-group variation (SCNT vs control). Nevertheless, in order to avoid confusion we have changed the term “indistinguishable” to “similar” throughout the manuscript:

“Both cell lines exhibited normal ESC morphology, expressed key stemness markers, were transcriptionally similar to control ESCs and differentiated into multiple cell-types in vitro and in vivo.”

The authors claim that the reprogrammed ESCs possess the potential to differentiate into multiple cell types would be much stronger if supported by data on the epigenetic status of the reprogrammed ESCs. Because the first step of nuclear reprogramming requires erasure of donor cell epigenetic patterns and re-establishment of embryonic epigenetic characteristics in the cloned embryo and because methylation is a major imprinting mechanism, the pluripotent potential claimed by the authors would be more convincing if supported by methylation data such as global methylation analysis. These data would not provide gene specific methylation status, but would provide data showing that global methylation is or is not similar to control IVF produced ESCs. A difference of 6% in global methylation as seen with the gene arrays would suggest that the reprogrammed ESCs are not indistinguishable from IVF-produced ESCs.

Global methylation analysis would require extraction of genomic DNA from control and experimental ESCs and somatic cells using standard methods (most likely this was done previously for microsatellite genotyping). Global methylation using nearest neighbor analysis is a straight-forward method that would require less than one week of bench work and data analysis.

We agree with the reviewer that analysing the global methylation status of the CRES, control and donor cell lines would be an interesting field for future research and we hope to use epigenetic analysis of CRES cells and individual SCNT primate blastocysts as the basis of a future SCNT project.

The relationship between the work in this paper and the authors' recently published paper needs to be clarified.

Our previous research was focused on analyzing the rhesus monkey SCNT methodology in order to create SCNT blastocysts, while this study describes the derivation and characterization of rhesus monkey SCNT ESC lines. We have now included direct references to our paper published in Human Reproduction and have added the following paragraphs to the discussion:

“We speculate that the impaired blastocyst formation rate following conventional SCNT in primates may result from one or more of the following factors: Hoechst 33342 and/or UV damage to the relatively transparent primate oocyte, Hoechst 33342/UV induced oocyte activation and/or MPF degradation, reaction of the residual Hoechst 33342 in cytoplasts with the introduced donor cell DNA thereby impairing reprogramming and/or Hoechst 33342 contact with mitochondrial DNA, thus reducing cytoplast mitochondrial function. Further research is required to define the critical factors and mechanisms determining developmental success following different spindle removal and SCNT methodologies.

Recognising the importance of high quality cytoplasts for the successful reprogramming, we have been seeking non-invasive approaches for spindle detection and removal. “Blind” enucleation techniques involving “squishing” 20 or “one step manipulation” (OSM) 21 are intrinsically faulty, at least in our hands, because they fail to enucleate all oocytes. In fact, our initial effort to derive ntESCs from blastocysts produced with the OSM protocol resulted in the isolation of parthenogenetic ESCs, almost certainly due to failed spindle removal. Fortunately, recent developments in high-performance imaging resulted in an Oosight™ spindle imaging system supporting non-invasive, rapid and highly efficient real time enucleation of primate oocytes.”

The manuscript needs clarification of gene expression array statistical methods including clarification of how “significantly different” gene expression is defined. Explanation of CV values for comparison of cell line transcriptome data. Additional evidence that the reprogrammed ESCs are indistinguishable from IVF-produced ESCs.

We used the standard MAS-5 (Microarray suite 5) statistical algorithms incorporated in the Affymetrix GCOS program to analyse statistically significant change between samples. We use the following text to explain this:

“MAS-5 statistical analysis was performed to calculate the signal log ratio (SLR) for each probe set comparison, and the gene expression fold-changes (FCs) between two samples were calculated from the SLR using the following formula: FC = (2SLR). The GCOS 1.4 MAS 5.0 software was used to calculate statistically significant differences in gene expression (P < 0.002) between samples.”