Feedback

Tell a friend

Print

Johns Hopkins’ Aravinda Chakravarti and Anirban Maitra, and Mahendra Rao at the National Institute on Aging discuss the use of high-resolution methods for quality control of cell lines

Scientists led by Dr. Aravinda Chakravarti, director of the McKusick-Nathans Institute of Genetic Medicine at Johns Hopkins Medical Center and Mahendra Rao, head of the Stem Cell Group in the Laboratory of Neuroscience at the National Institute on Aging, report that they have uncovered mutations in federally approved embryonic stem cell lines that have been cultured in the laboratory over extended periods of time. Identification of genetic abnormalities in these lines cast doubt on their safety and efficacy for therapeutic use and has many implications for biological studies on them.

“These mutations may be widespread in all kinds of cultured cells, but genetic changes are particularly important in embryonic stem cells,” said Chakravarti.“Everybody is hoping to use these cells in some therapeutic protocol one day, so we need to know exactly what the cells are, what kinds of genetic changes they have, and whether those changes make any biological difference or not.”

The use of embryonic stem cells has been a controversial issue since 2001, when the Bush administration restricted publicly funded embryonic stem cell research to what are now 22 existing cell lines, a move that many scientists believe stifles important advances in stem cell research.

The work, which was done largely using the Affymetrix 100K SNP and Mitochondrial Resequencing Arrays, was published in the September 4th online version of the journal Nature Genetics. Conventional methods for analyzing the genetic integrity of cell lines relied on low resolution methodologies such as karyotyping. “I think at least one of the things that we have shown is that karyotyping is not adequate,” said Anirban Maitra, assistant professor of pathology and oncology, at Johns Hopkins. “Some cell lines that we looked at in this paper were karyotyped and found to be ‘normal,’ yet we found localized regions of amplification and deletion, as well as epigenetic changes in these cells.”

Chakravarti, who sits on the scientific advisory board of Affymetrix, Maitra, and Rao recently spoke to AMB editor Rachel Shreter. The four discussed:

• The types of changes they found in ES cell lines and why those changes occur
• Why they chose a microarray-based approach to look at the quality of today’s federally approved stem cell lines
• Future research that will provide more information about the types of changes that accumulate in these and other cell lines and how to prevent them from becoming prevalent
• The importance of using the data wisely from new studies to influence federal research policies

Genetic alterations in ES cell lines
Shreter: You have been looking at late passage human embryonic stem cell lines obtained from federally approved providers and found a number of mutations in them. Are these cell lines more susceptible to mutation than others?

Chakravarti: I think we’re using fairly powerful technologies on what are basically old problems and seeing new things. I’m not convinced these cells lines are any more unstable than other cell lines. It is quite possible, and even likely, that since we’ve never done these kinds of experiments—looking at gene copy number and genome-wide methylation or sequencing specific genes—on other cell lines, that all cell lines suffer these kinds of changes, and that a cell line normally considered to be pure, unadulterated cell line, might not be that at all. In fact, it might be a population of very closely related cells with many genetic changes.

We would hope the large majority of those changes are benign and don’t affect the function of those cells. They’re probably mostly changes in non-coding DNA with just a small fraction in coding sequences for proteins and in their regulatory regions. Even then, there is probably just a small minority of those mutations that make a difference to the phenotype of these cells. Then it depends on how many cells in those cell lines have a particular kind of mutation and what the nature of the mutation really is.

Shreter: What do you think causes these genetic changes?

Chakravarti: Well, the basic answer is we don’t know. Mutations occur during cell division in every kind of cell. So I think that’s the cause. I don’t think embryonic stem cells or any other cell lines are particularly vulnerable to mutation, since we know that DNA replication is not perfect and that mutations will occur.

We’ve always assumed that in a given cell line, the cells that have these mutations are all different from one another, so no particular change overrules any other kind of change. I think our studies show that there are at least some genetic changes that have occurred in substantial number or have allowed the mutant cells to overtake some of the non-mutant cells. So some speculation has been made in the paper about the quality of these cell lines.

Rao: In nature, there’s a relative inefficiency in the replication process. Fidelity is never 100 percent. And if you have cells dividing for a long period—whether it’s in culture or in the body—a small number of them will accumulate mutations that may be in a critical region. The accumulation of mutations can be directly related to the number of times the cells have divided.

Maitra: Different culturing methods also lead to different rates of mutation. There are two established culture methods for ES cells—enzymatic separation and mechanical dissociation. There is some published data that suggests the enzymes used may have an adverse impact on the chromosomes.

In fact, in our limited panel, the results suggest that mechanically dissociated ES cells had a lower incidence of copy number alterations than those enzymatically dissociated.

It is important to note, however, that this dichotomy was erased when we took epigenetic changes, like promoter methylation, into consideration. Cells separated by either technique demonstrated evidence of methylation. The ultimate effect of the dissociation technique on stem cell stability is something that will need to be examined in greater detail in a larger panel of ES cells.

Minimizing mutations
Shreter: Is there any evidence that suggests that these changes can be slowed down, stopped, or even reversed as the cells are passaged?

Rao: Well, if you figure out a way to identify the cells that don’t have changes, you can select or reselect or subclone out the normal population. People clearly need to look at the cells and identify changes as early as they occur, because then the chance of selecting cells without those changes is much better. So careful monitoring is one way of making sure that you can propagate a line, even in the face of stochastic changes.

Chakravarti: Often when we do find multiple changes, they are distributed across various cells. I think the only reversion that could happen is a back mutation of exactly the same kind. That’s very unlikely with the kind of major changes we are talking about.

On the other hand, some cells may have had major chromosomal changes that we don’t see. Those cells have died off because the mutation was lethal to that cell. In this paper, I think the mutations we saw were all randomly occurring. Some were benign mutations that have simply been carried along with no functional consequence whatsoever. But some mutations could have been lost or given a somewhat small growth advantage to the cell. Since we always pick cells that grow well for the next generation in cell culture, it’s possible that we inadvertently applied some kind of selection to those with good growth properties.

Using an array-based approach
Shreter: Why did you choose to use oligonucleotide microarrays for this study?

Chakravarti: We chose an approach that was available to us and could sample the genome at very high density to look for particular kinds of mutations. Microarray technology is now at a stage where we can screen for a certain class of changes. Even then, I’m fairly sure we did not detect all of the relevant copy number changes. The technology is still evolving. The technologies will improve and there’s no doubt in my mind that we will find more changes, not only in these cell lines, but in other cell lines as well.

Maitra: The current gold standard used for many of the embryonic stem cells is conventional karyotyping. I think at least one of the things that we have shown is that karyotyping is not adequate. Some of cell lines that we looked at in this paper were karyotyped and found to be “normal,” yet we found localized regions of amplification and deletion, as well as epigenetic changes in these cells.

Rao: The logic we used when we discussed our approach came from Dr. Chakravarti’s and Anirban’s experience with what happens in cancer cells. When cells change, the changes can be in the genome, epigenetic changes, changes in mitochondria, or changes in the microRNA that regulates gene expression.

We needed to figure out which method would allow us to sample each one of these qualities of a cell. And we had to choose from existing technology for which there was expertise and there was a reasonable cost-benefit to us. Given the microarray expertise in Dr. Chakravarti and Dr. Maitra’s laboratories, we thought that using oligonucleotide microarrays was the best option right now.

This test certainly isn’t exhaustive and we will need to do things at a higher resolution. We have improved the resolution for looking at these cells by using newer technology, but the resolution needs to get better still.

Investigating microRNAs
Shreter: Dr. Rao, you mentioned microRNAs. Have you started looking at those and the role they may play in stem cell regulation?

Rao: Several people have started looking at that. It turns out that microRNAs seem to be specifically enriched, and specific subsets seem to be uniquely expressed in embryonic stem cells and other stem cell populations. So investigators have begun using a variety of strategies including arraying microRNAs, sequencing them by enriching for small-sized RNAs, or actually going in and looking at microRNA binding sites in highly expressed genes in a stem cell population. And all of those strategies have yielded converging results. It appears there are several different classes of microRNAs that regulate several different aspects of function, including regulation of gene expression, regulation of the promoter, and regulation of RNA degradation.

Maitra: There are several competing technologies out there that we can use to study microRNAs. Most of them are array-based. What differs is the detection technique being used—whether it’s radionucleotide-based detection, or it’s a biotin label. The “good” thing about microRNAs is that there are only about 200 known, although a larger number is suspected. That makes them easier to study than, say, mRNA transcripts.

Future directions
Shreter: Where you see your work going in the future?

Maitra: Well, we talked about the available technology and why it has given us an unprecedented view of the stem cell genome. That is clearly not the entire story and we’d like to take a more magnified look at what’s going on and elucidate on a global scale the sort of aberrations that these cultured cells have.

We’d also like to understand the functional implications of these mutations. Mahendra talked about looking at microRNAs and additional changes at the epigenetic level. We’ll be looking at those and we’d also like to play a contributory role in determining the benchmarks required to pronounce these cells stable or not, and useable or not.

Chakravarti: There’s still so much more nuts-and-bolts work to be done. There’s no doubt that we need to look at this issue at a much higher resolution. I think the technology for reading the Affymetrix array—the software, the bioinformatics of all of that—will need to keep improving. As arrays obtain higher density, our ability to detect subtle changes will depend on much, much better methodology.

As you add more oligonucleotides and your feature size goes down, you’ll need to address how the arrays can be read on different kinds of scanners, how the data can be filtered and how to interpret the data. It’s an entirely technological issue having nothing to do with biology, but it’s informed by biology. The majority of changes that we will need to look at and have to detect will be heterozygous changes, small deletions, or small copy number changes. We’re still not very good at detecting those because of a host of technical issues and those need to be addressed.

Influencing federal policy
Shreter: How would you ultimately like to see your work affect federal policy?

Rao: When Dr. Zerhouni went before the Senate, they asked him whether we need more cell lines. And he said, more cell lines could not hurt. I think that’s very true, and I think the data here suggest that it would be useful to have a lot more cell lines.

Chakravarti: It is important that we use this work, and many other kinds of work, to show that our existing cell lines are insufficient for us to do any biology, let alone contemplate therapeutic use. Although I think this field has tremendous amount of promise, it’s years away from any kind of therapeutic application.

There’s absolutely no doubt that this country is very divided over stem cell issues and we need to work towards a more informed consensus. At the same time, as scientists, we need to be much more honest about the current limitations of stem cells and what we still need to learn for clinical applications. I’m certainly not saying the federal government should stick to its policy. But I think we scientists also have to be a little bit more realistic about the time frame for applications.

FOR MORE INFORMATION
Contact Information Aravinda Chakravarti, Ph.D. Johns Hopkins University School of Medicine Institute of Genetic Medicine 733 N. Broadway Broadway Research Building, Suite 579 Baltimore, MD 21205 Anirban Maitra, MBBS Assistant Professor of Pathology and Oncology Affiliate, McKusick-Nathans Institute of Genetic Medicine The Sol Goldman Pancreatic Cancer Research Center Johns Hopkins University School of Medicine Ross Bldg 632 720 Rutland Ave Baltimore, MD 21205 Mahendra Rao, M.D., Ph.D. Head, Stem Cell Group National Institute on Aging National Institute of Health Laboratory of Neurosciences 5600 Nathan Shock Drive Baltimore, MD 21224	Upcoming Events Affymetrix online seminar with Anirban Maitra Monitoring Genetic Stability in Human Embryonic Stem Cells Monday, October 31, 2005 1:00 pm, Pacific Standard Time Register here Background Microarray applications in stem cell research Companies Affymetrix Inc. Organizations: Johns Hopkins University School of Medicine National Institute on Aging Further Reading Maitra A, Arking DE, Shivapurkar N, Ikeda M, Stastny V, Kassauei K, Sui G, Cutler DJ, Liu Y, Brimble SN, Noaksson K, Hyllner J, Schulz TC, Zeng X, Freed WJ, Colman A, Sartipy P, Matsui SI, Carpenter M, Gazdar AF, Rao M, Chakravarti A. Genomic Alterations in Cultured Human Embryonic Stem Cells Nature Genetics 2005; Sept.4 [Epub ahead of print]	Maitra A, Cohen Y, Gillespie SE, Mambo E, Fukushima N, Hoque MO, Shah N, Goggins M, Califano J, Sidransky D, Chakravarti A. The Human MitoChip: a high-throughput sequencing microarray for mitochondrial mutation detection. Genome Res. 2004; 14(5):812-9. Cutler DJ, Zwick ME, Carrasquillo MM, Yohn CT, Tobin KP, Kashuk C, Mathews DJ, Shah NA, Eichler EE, Warrington JA, Chakravarti A. High-throughput variation detection and genotyping using microarrays. Genome Res 2001; 11(11):1913-25. Noaksson K, Zoric N, Zeng X, Rao MS, Hyllner J, Semb H, Kubista M, Sartipy P. Monitoring differentiation of human embryonic stem cells using real-time PCR. Stem Cells 2005; Aug. 4 People Elias A. Zerhouni, MD Director, National Institutes of Health http://www.nih.gov/about/director/