This is a project based on the idea that micro RNAs (miRNAs) could be used to reprogram cancer cells into induced pluripotent stem (iPS) cells. The proposal seeks to generate micro RNA-induced pluripotent stem cells (mirPS cells) by driving expression of mir-302.
The proposed idea of using miRNAs to reprogram cancer cells is an interesting idea, particularly in the context of the mir-302 cluster, which the principal investigator (PI) indicates is highly expressed in oocytes and embryonic stem (ES) cells. However, one of the strongest disagreements with this proposal voiced by reviewers concerned the scientific rationale of utilizing cancer cells for reprogramming and subsequently for therapeutic applications in human. It is likely that the inherent genetic instability of cancer cells leads to undesired alterations, thus limiting the potential utility of the derived cells, even if such cells can be reprogrammed and pluripotency is demonstrated. In addition reviewers disagreed with the applicant’s notion that the use of mirPS cells is advantageous because they will bypass immunocompatibility issues faced by the use of human ES cell-derived tissues. iPS cells derived from normal somatic cells potentially have the same advantage without the risk of exposing patients to cancer cell-derived tissues. In addition, strong criticism was expressed about the statement that this approach will “immediately provide patient-friendly stem/differentiated cell lines useful for clinical trials”, as the PI claims.
In terms of feasibility of the proposal, the premise behind using mir-302 to reprogram cells appears to be sound. mir-302 is expressed at high levels in mouse oocytes and human ES cells and preliminary data supports that expression of the mir-302 family can induce ESC-like changes in a melanoma cell line. The major issue with the proposed experiments is that it is not clear that this approach will produce useful human cell lines.
The proposal to investigate the role the mir-302 cluster potentially plays in reprogramming cells is interesting from the standpoint of miRNA biology. However, the rationale for using tumor cells to generate pluripotent cell lines for regenerative medicine is not well justified. It is not likely that new, useful human ES cell lines will be generated from the proposed experiments.
Reviewer One Comments
The idea of using miRNAs to reprogram somatic cells is an interesting idea, particularly in the context of the mir302 cluster, which the principal investigator (PI) indicates is highly expressed in oocytes and embryonic stem (ES) cells. The stated rationale is to develop methods to provide “patient friendly” iPS cells from cancer cells to treat human diseases. There may be potential to use miRNA technology for tumor therapies. It is however very unclear why it would ever be a good idea to use cancer cells to generate tissue that would be put back into patients. The likelihood of alterations from genetic instability of cancer cells seems a formidable challenge, even if they do demonstrate features of pluripotency.
The PI has generated cell lines from human cancer cell lines (breast, colon, prostate) that express stem cell markers, by forced expression of mir302, and he/she refers to these as mirPS. The first Aim is to compare gene expression and epigenetic modification patterns of human embryonic stem cells (hESCs), mirPS, and the parental cancer cell lines used to generate mirPS. This will entail microarrays for transcript profiling, methylation status of promoters for stemness genes, and histone acetylation patterns. These are feasible experiments, although data for hESCs is already widely available.
Experiments in the second Aim will assess pluripotency of the mirPS lines. Karyotype will be evaluated, and it would be interesting to know the karyotype of the starting cell lines. Telomerase activity and differentiation capacity will be tested in vitro, and following IP injection into SCID-beige mice to generate teratomas. Another set of injections employs pseudo-pregnant donor mice, although the point was not made clear for what purpose.
The third Aim is to attempt to develop feeder free differentiation protocols using mirPS cells, focused on dopamine neurons and pancreatic beta cells. Although not explained in detail, it appears that the idea would be to use cancer cells to treat diabetes, assuming perhaps that the patient has both problems?
Responsiveness to RFA:
The proposal seeks to generate a novel type of pluripotent cell line, using expression of mir302 to generate mirPS. The rationale of using tumor cells to generate pluripotent cell lines for regenerative medicine is not well justified. The PI has expertise in prostate cancer, and this focus might be a good use for mir-based therapies.
Reviewer Two Comments
This proposal will use the mir-302 cluster to reprogram cancer cells. The rationale behind these experiments is not clear. It is very unclear why cancer cells are being used for these experiments. In Aim 3, the PI states that the use of mirPS cancer cells will bypass immuno-compatibilty issues but so would performing the proposed experiments in a non-cancer cell line. In addition, I strongly disagree that this approach will “immediately provide patient-friendly stem/differentiated cell lines useful for clinical trials”, as the PI claims.
The premise behind using mir-302 to reprogram cells appears to be sound. Mir-302 is expressed at high levels in mouse oocytes and human ES cells and preliminary data supports that mis-expression of the mir-302 family can induce ESC-like changes in the Colo cell line. The major issue with the proposed experiments is that it is not clear that this approach will produce useful human cell lines.
There are a number of advantages cited for using mir-302 expression to induce reprogramming over the expression of transcription factors. For example, one is that some of the transcription factors currently used are oncogenes. There is no evidence that mir-302 is not ALSO an oncogene. A number of miRNAs have been shown to function as oncogenes. The PI presents no evidence that the mir-302 family does not function as an oncogene and the fact that forced expression of the mir-302 family appears to at least partially reprogram cells suggests that in fact the mir-302 family may be an oncogene. If mir-302 is an oncogene then there does not appear to be an advantage, and possible a number of disadvantages such as the ability of a single miRNA to activate hundreds of target genes, in mis-expressing this family for use in reprogramming.
The PI states clearly in the abstract that another advantage of using miRNAs to reprogram cells is that the expression of transcription factors currently requires retrovial vectors. However, the method used in the proposal to reprogram using miRNAs ALSO uses a retroviral vector (Fig 1). Since both approaches use retroviral vectors I fail to see the proposed advantage.
The very strong CMV promoter was used to drive expression of the mir-302 cluster for reprogramming. Reprogrammed cells will then be differentiated into a number of cell types. There is no discussion/experiments designed to address the continued expression of the mir-302 family throughout the entire life of the cells. The CMV promoter will continue to drive strong expression of the mir-302 cluster in cells that have been differentiated in vitro. While the use of controlled media in vitro may be able to overcome forced expression of mir-302 and keep the reprogrammed cells in the desired cell fate, it is likely that when in vivo at least some of the cells will dedifferentiate. Dedifferentiated cells that continue to express high amounts of mir-302 could result in cancer. This is especially true if the cells, as proposed by the PI, are cancerous to begin with.
Some less major issues:
In Fig 2, it is stated that in the “absence of proper guidance under feeder-free conditions, these transfected cells tend to differentiate into neuron-like cell types”. Is this for all three cell lines examined? The PI states in the intro that they have evidence that the mir-302 family tightly shuts down neuron-specific cell differentiation. The data in this figure is a bit confusing and appears to contradict this hypothesis.
From the data presented in fig6, it is not clear that the Colo+mir-302 cells are pluripotent. There is an arrow in some of the panels but I have no clue what it is pointing to.
There is no fig4.
In all three aims the PI states that there will be “no expected problems”. This statement is a bit naïve and suggests that the proposed experiments have not been thought about in detail.
Responsiveness to RFA:
The proposal to investigate the role the mir-302 cluster potentially plays in reprogramming cells is interesting from the standpoint of miRNA biology. It is not likely that new, useful hES cell lines will be generated from the proposed experiments.
Reviewer Three Comments
This proposal targets the generation of iPS cells from cancer cells via miRNA gene silencing. While an interesting hypothesis and not with theoretical merit, the resulting cell lines would not be useable as patient friendly iPS cells for transplantation back into human.
This research plan suffers from a number of flaws too numerous to list here, most notably the usefulness of any resultant iPS cells.
Responsiveness to RFA:
The generation of a cell line that contributes to all three germ layers is unlikely given the current proposal.