Archive ScienceWatch



Croce Ohio State’s Carlo M. Croce - A Macro View of MicroRNA
Featured Science Watch® Newsletter Interview
(Science Watch, 19[2]: 3-4, March/April 2008)

RNA genes of only tens or hundreds of nucleotides have been called the biological equivalent of dark matter— "all around us but almost escaping detection." These genes are found in that portion of DNA that doesn’t code for proteins and so was commonly known as "junk DNA," which explains why they received so little research attention until the late 1990s. Nowadays, however, research into these so-called microRNAs (miRNAs) and their role in cellular function and dysfunction—notably, in the latter case, cancer—has blossomed into one of the hottest fields of biology and medicine.

Leading the pursuit of miRNAs as fundamental players in the cancer process is the Ohio State University cancer geneticist Carlo Croce, who reported in November of 2002 that chronic lymphocytic leukemia (CLL), the most common human leukemia, appears to be caused by the loss of two miRNA genes. These two miRNAs target the BCL2 oncogene.

Croce’s 2002 paper reporting this discovery has now been cited more than 300 times (see adjoining table, paper #5). More recently, a 2005 report by Croce and colleagues on this topic promptly found its place in the Hot Papers database, garnering more than 170 citations in just over two years (A. Cimmino, et al., PNAS, 102[39]: 13944-9, 2005). This is just one of 10 reports published by Croce and colleagues in the last two years that registered as Hot Papers during 2007, a total that earned him a spot on the annual list of "hot" authors in this issue’s lead story.

Indeed, Croce’s research on miRNAs has made him, at this writing, the seventh most-cited author in biology in the latest update to Thompson Scientific’s Essential Science IndicatorsSM database, based on papers published and cited over the last decade. Since 1991, Croce has published four articles that have each garnered more than 500 citations, and over 60 papers exceeding 100 citations each.

Croce, 62, did his undergraduate and medical studies at the University of Rome, where he graduated summa cum laude in 1969. He then moved to the Wistar Institute in Philadelphia and pursued his research in cancer genetics there for much of the next 20 years. In 1988, Croce became director of the Fels Institute for Cancer Research at the Temple University School of Medicine. Three years later, he moved to Thomas Jefferson University to head up the Kimmel Cancer Institute/Kimmel Cancer Center. In 2004, Croce relocated to Ohio State University, Columbus, where he is now chair of the Department of Molecular Virology, Immunology, and Medical Genetics, and director of the Human Cancer Genetics Program and the Institute of Genetics.

Croce spoke to Science Watch® from his office in Columbus.

When was the first miRNA gene discovered, and how did the field evolve over its first few years?

The first microRNA was discovered by Victor Ambros at the Dartmouth Medical School in 1993. He discovered this gene known as lin-4, which is important for the development of the worm, C. elegans. Then, other researchers—Gary Ruvkun at Harvard, most prominently—found other miRNA, such as let-7. After that there was a pretty long gap. Very few people were interested in miRNA until 1998, when it became clear that small interfering RNA, or siRNA, was taking advantage of the same pathway that exists for miRNA.

What do miRNAs actually do?

Highly Cited Papers by Carlo M. Croce and Colleagues, Published Since 1996
(Ranked by total citations)
Rank     Paper Cites
1 M. Ohta, et al., "The FHIT gene, spanning the chromosome 3p14.2 fragile site acid renal carcinoma-associated t(3;8) breakpoint, is abnormal in digestive tract cancers," Cell, 84(4): 587-97, 1996. 707
2 S. Haldar, J. Chintapalli, C.M. Croce, "Taxol induces bcl-2 phosphorylation and death of prostate cancer cells," Cancer Res., 56(6): 1253-5, 1996. 475
3 G. Sozzi, et al., "The FHIT gene at 3p14.2 is abnormal in lung cancer," Cell, 85(1): 17-26, 1996. 435
4 S. Haldar, A. Basu, C.M. Croce, "Bcl2 is the guardian of microtubule integrity," Cancer Res., 57(2): 229-33, 1997. 420
5 G.A. Calin, et al., "Frequent deletions and down-regulation of micro-RNA genes MiR15 and MiR16 at 13q14 in chronic lymphocytic leukemia," PNAS, 99(24): 15524-9, 2002. 332
SOURCE: Thomson Scientific
Web of Science®

They bind to the 3’ untranslated regions of RNA and cause a block of translation and some degree of degradation. In other words, they are regulators of gene expression. But it was only in 1998, with the discovery of siRNA, that this became clear. siRNA uses cellular mechanisms for silencing that also exist for miRNA. So at this time more people started to get interested. In 2000, homologues of miRNA were found in Drosophila. And then in 2001, several laboratories found that, in fact, this miRNA had homologues in rats, mice, and humans.

And this was in so-called junk DNA?

Well, it was called junk. Clearly, it’s not.

When did you make the first connection between miRNA and cancer?

That was in 2002. For a long time, we’d been taking advantage of specific chromosomal alterations to identify and characterize oncogenes or tumor suppressor genes that are involved in a large variety of human cancers—from acute leukemia to solid tumors, like lung cancer. We had characterized many of them, but then, in the early 1990s, we tried to identify the tumor suppressor on chromosome 13 in band q14 that was related to chronic lymphocytic leukemia. This region is deleted in about 50% of CLL cases.

For seven years we tried to find this tumor suppressor gene and we couldn’t do it, although we characterized all the genes in this region that are preferentially lost. Finally, by using CLL cells carrying the translocation and this small deletion, we were able to find that the culprits were not protein-coding genes, but two miRNA genes, called miR-15 and miR-16.

As it turns out, these miRNA genes are lost, or expressed at a very low level, in CLL. We showed that they are knocked out or knocked down in 68% of CLL cases. And this was the first link between miRNA and cancer. Then we mapped all known miRNA genes, and found that many mapped to regions involved in rearrangements in cancer. We then developed the first chip to detect the global expression of miRNA in normal tissue versus cancer tissue.

Subsequently, we started making the signatures for specific human cancers, looking at the function of various miRNAs to see if they work as oncogenes or tumor suppressor genes. We also developed the first genetic mice model. We deregulated a mouse miRNA gene, miR-155, in mice and observed that those mice developed cancer. That paper was published two years ago.

The mice developed leukemia?

Yes. After we deregulated one miRNA, 100% of those mice developed an acute leukemia.

Why do you think miRNA escaped detection or notice until as late as 1993?

Because people were affected by dogma. The dogma was that RNA exists to make proteins. So all this RNA found to be non-coding was thought to be junk.

When you discovered that these two miRNA genes were the culprits in CLL, was it obvious or did it take some convincing that you hadn’t made a mistake?

When I realized that the only things that could be involved in CLL were these two miRNAs, it was a revelation. I asked myself, how could we have been so dumb? Again, it really was a revelation, because it indicated that another totally different class of genes could play a major role in cancer.

How much of your laboratory time and resources are now devoted to miRNA research?

I have a pretty large laboratory, and I’d say 70% of my people work on miRNA.

How would you describe the state of the miRNA research universe circa early 2008?

The big picture is that these miRNA can certainly be used for cancer diagnostics and prognostics. And not only for cancer—it’s likely that many other diseases are also due to miRNA dysfunction. I think we’ll soon see miRNA used as therapeutic drug. MiRNA can get into cells, so we can use miRNAS that are lost in cancer as therapeutic agents—we can put them back. And if they’re over-expressed in cancer, we can use anti-miRNAs—small molecules—that can get into cells and regulate them. In a year or two we will probably see the development of miRNA-based therapies.

Are there biotech or pharmaceutical companies already pursuing the idea of miRNA therapeutics?

Not yet, but they are certainly starting to think about it.

How many miRNA genes are there likely to be in humans?

About 450 have been discovered so far, and there’s another 700 that have been predicted. Some of those will turn out not to be correct.

What are the predictions based on?

On bioinformatics.

And how are these predictions generated using bioinformatics?

You look at the structure of RNA and how they’re made. Then you keep all this in mind and look at the gene areas and the DNA and see which DNA might possibly generate this mature RNA. That gives you the prediction, which you then have to validate.

What makes you think miRNA might play a role in diseases other than cancer?

We know many diseases which are genetic, as shown by twin studies and family studies, but we can’t find the genes involved. For example, bipolar disease and schizophrenia are genetic diseases. Twin studies and family studies indicate that alterations in genes are the cause of these diseases. But you can’t even map them. Nobody has found them. So this suggests that they might be caused by alterations in miRNAs, because that’s precisely the scenario you would see if this were the case. And so it’s likely that one day we’ll find that these diseases are due to genetic alterations of miRNAs or defects in the processing of miRNAs. Alzheimer’s is another one of these cases where it’s plausible that miRNAs are the culprit.

So the failure to identify the genes responsible suggests to you that miRNA is the answer?

Yes. How did we find the involvement of miR-15 and miR-16 in chronic lymphocytic leukemia? We looked for the protein-coding gene and couldn’t find it. The reason we couldn’t find it is because this disease is caused by miRNAs—in this case, the loss of the genes. So if we apply the same line of reasoning to these other diseases, then the fact that a lot of very smart people looked for many, many years to find the protein-coding genes and failed suggests that the disease is due to alterations not in protein-coding genes, such as miRNA genes.

Are there other types of RNA that might play a role in cancer or other diseases?

We had a paper in the September 2007 issue of Cancer Cell reporting that another large family of non-coding RNAs, what we called ultraconserved, non-coding RNAs, are expressed in a lot of different tissues, and are differentially expressed in different tissues, and their pattern of expression is altered in cancer (G.A. Calin, et al., Cancer Cell, 12[3]: 215-29, 2007). The RNAs are much bigger than miRNA, from 200 nucleotides to 1.8 kilobases.

These RNAs are conserved 100% between rats, mice, and humans, hence the name. We find that these ultraconserved, non-coding RNAs are dysregulated in cancer. It’s also possible that miRNAs might regulate not only protein-coding RNAs but other non-coding RNAs as well.

Give us a prediction for where miRNA research will take us in the next five years.

It will be a revolution. We will have drugs based on this, and a lot of novel diagnostic and prognostic markers will be developed. The great opportunity will be in therapy. We will understand the diseases much better, including cancer, heart disease, and many other diseases, and we’ll be able to use miRNA or anti-miRNA to go in and perhaps reverse or alter the disease process.


2008 : April 2008 - Author Commentaries : Ohio State’s Carlo M. A Macro View of MicroRNA