June 15, 2011 (Vol. 31, No. 12)

Ellyn Kerr

Gene Regulators Show Promise as Biomarkers and Potential Therapeutics

MicroRNAs—as an entirely new class of gene-expression regulators (discovered in humans in 2001)—have elicited widespread interest. These noncoding, single-stranded, 20 to 25 nucleotide RNAs regulate gene expression through sequence-specific mRNA hybridization (usually at the 3´ untranslated region) and have been shown either to block translation or steer the degradation of mRNAs.

But because miRNA function does not require perfect binding complementarity to mRNA targets, a single miRNA may regulate multiple mRNAs. This, coupled with an integral involvement in several biological processes (including immune responses, cell-cycle control, stem cell differentiation, and metabolism), as well as their sheer collective number (as many as 1,100 identified in humans to date), renders miRNA both a rich and confounding source of study.

A broad range of challenges and innovations in miRNA research were presented at CHI’s recent conference on microRNA in Human Disease and Development.

Analysis & Insight: Are Early Clinical Successes Enough to Bring RNAi Back from the Brink?

A related and relatively mature field to miRNAs is siRNA, but researchers are still trying to overcome developmental challenges. There’s been some recent good news, in terms of both clinical data and litigation settlements. Get the details here.

Extracellular miRNA

Kai Wang, Ph.D., senior research scientist at the Institute for Systems Biology, discussed extracellular miRNA as a new source of biomarkers. Dr. Wang and colleagues in David Galas’ lab are studying many aspects of miRNA, both functional and disease-correlative. By profiling both miRNA and mRNA expression levels in numerous body fluids, including breast milk, urine, and plasma, they determined various miRNA signatures (profiles) discretely associated with various biological states (e.g., pregnancy, presence of disease, stage of disease).

The identification of stable miRNAs in extracellular environments suggests a biological function that may relate to cell-cell communication. Dr. Wang’s research aims to understand the potential of miRNAs as biomarkers, as well as their mechanisms of function, intra- and extracellularly, through study of human samples and animal models.

The challenges, as he acknowledged in his presentation, are current limitations in miRNA measurement and assessment. “When we evaluate miRNA measurement data, we need to be extremely judicious.” Using different platforms may yield significantly different results, which can complicate data verification and validation.

It is commonly acknowledged, for instance, that fluorescence- or other colorimetric-based microarray data can vary from that obtained with the same samples using qPCR, Dr. Wang noted. The imperative is, then, as much as possible, to use the same platform within a study and to be aware of possible differences between measurement methods, he advised.

Automated Multiplexing

Employing multiplexed assays to identify both miRNA and protein markers can strengthen conclusions about microarray data. Lorenzo Sempere, Ph.D., research assistant professor of medicine at Dartmouth Medical School, has adopted a multiplexing strategy combining in situ hybridization and immunohistochemistry for co-detecting, respectively, miRNA and clinically relevant protein markers at single-cell resolution.

Standard microarrays can fail to distinguish “altered miRNA expression within the cancer cell compartment or other elements of the tumor microenvironment; for example, upregulation of a miRNA in cancer cells or immune cells may each have etiological relevance and clinical significance, but obviously indicate vastly different biological processes,” he said.

“In addition, the ratio of cancer cells to stromal cells in different tumors can vary substantially, confounding interpretation of results. A tumor sample with a high percentage of cancer cells expressing a particular miRNA at low levels may yield similar results as a tumor sample with a low percentage of cancer cells expressing that same miRNA at very high levels.”

Dr. Sempere has developed multiplexed in situ fluorescence assays which he described as appropriate for current diagnostic workflows in typical clinical pathology labs. His team is implementing a fully automated pipeline of marker detection and computer-assisted analysis for “morphology-driven tissue slide-based assays,” applying patented Dartmouth miRNA biomarkers and detection methods. His goal is to “accurately and reproducibly” determine the role of miRNA-mediated processes in different cellular compartments.

Dr. Sempere has applied in-house synthesized fluorochromes for detection of other RNA markers to assess RNA quality and integrity, and protein markers to characterize particular cellular components associated with specific cell identities. He has used the multiplexed assays to localize expression of miR-21 and miR-155 (putative oncogenes known to be upregulated in cancer), to cancer cells and reactive stroma, and to immune cells in solid tumors, respectively. Such characterizations will serve to “increase the diagnostic power” of aberrant miRNA expression analyses, he said.

Therapeutic Targets

Expanded characterization of miRNAs is advancing their potential as therapeutic targets, not just as biomarkers. Eva Hernando, Ph.D., assistant professor in NYU School of Medicine’s department of pathology, is assessing the diagnostic and prognostic aspects of miRNA, as well as therapeutic capacity, with a focus on melanoma metastasis. Her work is directed at using miRNAs as biomarkers to discriminate patients with primary melanomas at higher risk of recurrence, and then to be able to target relevant miRNAs, either by upregulating those underexpressed in the cancer state or functionally ablating those with oncogenic effects.

At the CHI meeting, Dr. Hernando presented results from miRNA expression arrays that associated increased miR-30d expression with advancing stages of melanoma, increased cell invasion in vitro, and with melanoma metastasis in vivo.

She too, acknowledged limitations in assessing miRNA data. Use of miRNA arrays versus qRT-PCR allows analysis of greater numbers of miRNAs but with loss in sensitivity. Array data must, therefore, always be validated, “so we perform independent validation, by qPCR, of any miRNA hits we see differentially expressed that are associated with metastases or other clinical feature of interest,” Dr. Hernando said.

Her lab is working with Regulus Therapeutics, the branch of Alnylam Pharmaceuticals focused on miRNAs, to explore miRNAs’ therapeutic targetability. Dr. Hernando’s lab recently published proof-of-principle results in a mouse model of melanoma liver metastasis; synthesized anti-miR-182 oligos were shown to downregulate tumors and upregulate several miR-182 targets, suggesting the potential of miRNA as an antisense-drug target. She observed that efficacy in the liver has been well demonstrated but that companies must improve delivery methods for efficacy in other tissues (an area for future research).

Knockout Approach

Knowledge of the precursor steps in miRNA biogenesis affords the opportunity of creating full miRNA knockouts, which Robert Blelloch, M.D., Ph.D., associate professor in the urology department at the University of California San Francisco School of Medicine, is using to characterize prostate cancer. Cassandra Belair, Ph.D., a researcher in Dr. Blelloch’s lab, presented data on miRNA signatures and their roles in prostate cancer progression.

miRNAs in plasma and other body fluids provide a readout of the disease state of a patient. The Blelloch lab has sought to further develop “highly sensitive and accurate assays” for profiling miRNAs in the sera of patients with prostate cancer, using knockouts in the biogenesis pathway to follow the accuracy of the assays. For example, removal of Dgcr8 should result in the loss of all canonical miRNAs (referring to a particular miRNA biogenesis pathways), while the loss of Dicer should result in the loss of both canonical and noncanonical miRNAs.

Multiplex PCR quantitation techniques used in commonly available kits can be “remarkably inaccurate,” even showing upregulation in the knockout cells, according to the research team. Through critical modifications, they “improved on these techniques to render an accurate, high-throughput process to quantify all miRNAs.” The process has been adopted to a nanofluidics PCR platform developed by Fluidigm designed for very low-volume reactions, thereby conserving both patient samples and valuable assay reagents.

In proof-of-principle experiments, these methods were reportedly shown to be “highly sensitive, easily detecting miRNAs in patient sera.” Importantly, the lab was able to identify miRNA signatures that correlate with patient risk of progression as defined by the commonly used metric, the Cancer or Prostate Risk Assessment Score (CAPRA).

Unlike many other studies, samples were taken from patients who had not yet received prior treatment; results, therefore, reflected biomarkers of disease risk at the stage when decisions regarding interventions must yet be made. Based on these findings, the Blelloch lab is extending its studies to additional patient populations with the aim of providing a new component to the risk metrics for prostate and other cancers.

The research team has a main focus on the functional roles of miRNAs in development and in disease. The Dgcr8 knockout tool is employed as a starting point for many of their studies. By initially removing all canonical miRNAs, study of individual miRNAs in isolation is facilitated. This approach has been successfully applied by the lab with embryonic stem cells, and in prostate cancer (with preliminary results obtained in a commonly used mouse model of the disease). Data showed that while miRNAs may not be required at the earliest stages of disease, they are essential for progression.

The Blelloch lab is now using the animal model to uncover relevant individual miRNAs and systematically dissecting downstream pathways regulated by those miRNAs. Such studies will provide a systems, or network, view of genes important in prostate cancer patients, including nodes that can then be used to direct small molecule development.


Expansion of basal and luminal cells in a mouse model of prostate cancer: Samples of normal and prostate cancer were stained for cytokeratins specific for basal (CK5 in green) and luminal (CK18 in red) cells. Nuclei are labeled in blue. [UCSF]

Natural Products

Fazlul H. Sarkar, Ph.D., professor of pathology at Karmanos Cancer Institute at Wayne State University, is profiling miRNA signatures associated with various cancers and other diseases and studying means of regulating miRNA activity by natural agents with known anticancer effects.

“We are looking at active compounds in dietary agents involved in either destroying miRNAs with oncogenic activity or augmenting the expression of miRNAs that are downregulated in diseases,” Dr. Sarkar said. The complete mechanisms of activity have yet to be characterized, but he has hypothesized that demethylation of selective promoter sites is involved in reexpression of miRNAs that suppress oncogenesis.

His lab is also exploring cancer recurrence stemming from cells that have survived in patients through past treatment regimens. “Lab findings, including our own, have shown two distinct features of such cells, including one called epithelial-to-mesenchymal transition, or EMT.

“Most epithelial tumors are epithelial-cell tumors that are morphologically and genotypically epithelial; when cancer is survived through a therapeutic insult, such cells can change their phenotype and become mesenchymal,” Dr. Sarkar explained, whereby they can opportunistically invade the vasculature to relocate to eventual metastatic sites within the body. Such cells also acquire cancer stem cell characteristics.

Dr. Sarkar’s lab is exploring the miRNAs involved in cancer recurrence and their responses to such natural agents as indoles, isoflavones, and curcuminoids. The work is yet at proof-of-principle stage. Three Phase II trials of isoflavones in prostate and breast cancer showed positive results, with one study indicating a reduction in radiation-induced toxicity.

Previous Phase I trials in prostate cancer led to a Phase II study in patients with newly diagnosed cancer. A protocol has now been completed for a Phase II trial in patients with triple-negative breast cancer (negative for estrogen, progesterone, and Her2 receptors), offering an option for a disease clearly nonresponsive to conventional hormonal treatments.

Beyond the patient compliance that blood-test diagnostics permit, plasma-based diagnostic/prognostic assessments of miRNA avoid the problems of isolating from blood serum (whereby coagulation of proteins, with which miRNAs associate, can otherwise reduce yields and thus erroneously skew conclusions about miRNA levels in the samples). Ultimately, advances in miRNA R&D may open treatment options that drastically affect prognostic outcomes.

As proof of principle, Dr. Sarkar’s laboratory has shown that the expression of specific miRNAs whose expression is lost in aggressive cancer cells (miR-200 and let7) can be upregulated by these natural agents, opening doors for novel cancer therapies.

“We and others have shown that the expression of miRNA can be prognostically related to patient survival. In pancreatic tumors, for instance, median survival can be as short as six to eight months, whereas certain patients exhibiting low expression of miR-21 have a median survival of approximately 43 months.” Dr. Sarkar stated that the application of miRNA in cancer therapy is still in its infancy, and he predicted rapid progress going forward, to the benefit of patients.


EMT, CSC, and a few specific genes such as p53, nanog, and Lin28B have a role in causing tumor aggressiveness. Loss of miR-200 and let-7 also plays an important role, although both can be re-activated by natural agents, which will cause tumor regression. Tumor aggressiveness could also be due to overexpression of miR-21, but expression could be inhibited by natural agents, which will result in the killing of tumor cells. [Wayne State University]

Previous articleScientists Find Ovarian Cancer Cells Barge Through Mesothelium to Invade New Organs
Next articleAre Early Clinical Successes Enough to Bring RNAi Back from the Brink?