Chunxiao Ma

“A thing serving as a standard” —-This is the definition of a marker. As a graduate student, the concept of a
“marker” permeates every corner of my research life. When I
made our zebrafish video, I was very surprised to learn that we can make a green tumor and use it as a tool to monitor the shrinkage of tumor tissue. A related question then came to mind. Do we have similar tools to monitor tumor shrinkage in human cancer? This got me interested in searching for a biomarker of human cancer.

The cancer biomarker is a headache for scientists interested in cancer drug research. Years ago, when high-throughput techniques were only a dream, scientists wanted to find a molecule that can distinguish cancer cells from healthy cells. However, all the efforts seemed futile because cancer cells developed from healthy cells and thus were difficult to target.

The invention of microarray techniques provided new hope to the discovery process. It seems many biological materials (DNA, RNA, proteins) can be used for microarray analysis so long as one has appropriate ways to find the target. So mRNA, protein, and miRNA microarray profiles, which I will introduce in this blog, became favored techniques in approaches to find cancer biomarkers.

The biggest disadvantage of mRNA and protein microarray profiles comes from their large population, which makes a target difficult to detect. In the case of mRNA, for example, there are at least 22,000 mRNAs that can be used in the classifying procedure. Another problem is that the results from mRNA profiles are not always consistent with those from protein profiles, even when they come from the same tissue, indicating complications that lead to a high level of background noise [1]. One can use ribosome profiling to confirm that the microarray contains only the mRNA being transcribed, but the profile is not very sensitive or reproducible.

The newly discovered miRNA shows an advantage in distinguishing cancer types. miRNAs are a group of short non-coding RNAs that negatively regulate gene expression at the post-translational and/or translational level. The innate advantage of using an miRNA profile is that there are only about 200 miRNA discovered in the cell [2]. A number of research articles show that a specific miRNA or a cluster of miRNA influences each stage of cancer development [3]. In 2005, a group of scientists completed the profiling of miRNA from different cancer tissues and announced that miRNA expression profiles could serve as a classifier of human cancers [4].

miRNA profiling techniques; rectangular bars = miRNA. Single bar is the miRNA probe. Double bar is the miRNA probe hybridized with target miRNA. SAPE is used to detect double-stranded miRNA

This group first improved the specificity of profiling by using beads as a scaffold for hybridization (see flow chart). First, oligonucleotide-capture probes that are complementary to miRNA of interest are associated with beads containing two fluorophores. Then, purified miRNA is amplified and hybridized to the beads. After staining with streptavidin-phycoerythrin (SAPE), the beads are analyzed by flow cytometry using two lasers to quantify the levels of miRNA.

The results showed that almost all of the miRNA expression levels differed in the >20 different cancer cell types tested. Moreover, hierarchical clustering of the samples using miRNA profiles paralleled the developmental origins of the tissues. In other words, the expression of miRNA could be separated into different groups that corresponded to the origin of the cell type, which is particularly important when considering metastasizing tumors. Therefore, we conclude that miRNA expression patterns encode the developmental history of human cancer.

Clustering based on miRNA profile: red = cancer tissue, blue=normal tissue, green = cancer cell lines.

The next step is to compare the differences in the profiles of cancer cells and healthy cells. The statistical data show that most of the miRNA is down-regulated in cancer cells and tumor tissue. As a good biomarker should be able to distinguish the stage of the disease, further experiments were carried out to explore the function of miRNA in cellular differentiation. The results were the same in cells undergoing differentiation and cells that were chemically-induced for differentiation (see figure below). The data are important because they show that the technique can distinguish between differentiating and non-differentiating cells.

miRNA profiling

Because previous research shows that the abrogation of differentiation is a hallmark of cancer cells, this result shows that global changes in miRNA expression are associated with differentiation.

Finally, the miRNA expression profiles were used as a diagnostic tool of cells from primary tumors. Although the seventeen tissue samples came from poorly differentiated tumors, results from miRNA profiling correctly identified the cell type for twelve of the seventeen samples, a much higher success rate than than profiles that used mRNA (1/17) [5]. This result further supports the idea of using miRNA profiling as a biomarker for cancer diagnostics.

In summary, research into cancer biomarkers is an inviting aspect of cancer therapeutics, although to date, no uniform, official biomarker has been reported. In this blog, I introduce a new addition to this field, that of miRNA profiling. Much more research is needed, however, to move the technique from the laboratory to the clinic.

References:

1. Guo, H., Ingolia, N., Weissman, J., & Bartel, D. (2010). Mammalian microRNAs predominantly act to decrease target mRNA levels Nature, 466 (7308), 835-840 DOI: 10.1038/nature09267
2. Sato, F., Tsuchiya, S., Meltzer, S., & Shimizu, K. (2011). MicroRNAs and epigenetics FEBS Journal, 278 (10), 1598-1609 DOI: 10.1111/j.1742-4658.2011.08089.x
3. Liu, J., Zheng, M., Tang, Y., Liang, X., & Yang, Q. (2011). microRNAs, an active and versatile group in cancers International Journal of Oral Science, 3 (4), 165-175 DOI: 10.4248/IJOS11063
4. Lu, J., Getz, G., Miska, E., Alvarez-Saavedra, E., Lamb, J., Peck, D., Sweet-Cordero, A., Ebert, B., Mak, R., Ferrando, A., Downing, J., Jacks, T., Horvitz, H., & Golub, T. (2005). MicroRNA expression profiles classify human cancers Nature, 435 (7043), 834-838 DOI: 10.1038/nature03702
5. Ramaswamy, S. (2001). Multiclass cancer diagnosis using tumor gene expression signatures Proceedings of the National Academy of Sciences, 98 (26), 15149-15154 DOI: 10.1073/pnas.211566398