November 15, 2012 (Vol. 32, No. 20)

Gene expression profiling is considered the most direct approach in monitoring transcriptional and translational changes in cells and tissues.

Yet, in the earlier days, this technology was associated with great challenges. Classical gene expression assays made use of reporter genes, blotting techniques, and RT-PCR assays that were generally labor-intensive and time-consuming. A couple of decades later, multiplex approaches to gene expression allowed scientists to perform serial analysis of transcripts and gene copy number and reassembly of DNA segments into a contiguous path that could cover an entire genome.

Unfortunately, these large-scale approaches are usually associated with massive signal-to-noise ratio and are also extremely expensive to construct in the laboratory. Reports have also shown that arrays often come with elusive start and stop sites, decreasing the chances of generating a comprehensive analysis of the entire transcriptome.

According to Nalini Raghavachari, Ph.D., core director of the genetics and development biology center of the NIH Lung, Heart and Blood Institute (NHLBI), comparative RNA-Seq of different cell lineages has helped them understand pathways in cell differentiation and self-renewal. Their research program initially focused on using the high-density Affymetrix Human Exon 1.0 ST array platform for monitoring changes in the expression levels of genes and alternative splicing events of hematopoietic cells, both at the gene and exon levels.

“While conducting experiments on the array platform, next-gen sequencing was gaining momentum and when the opportunity came up for analyzing the same set of samples by RNA-Seq using SoLiD sequencing, we took it up and did a comparative analysis of both platforms,” explained Dr. Raghavachari.

She was one of a number of scientists who discussed novel approaches to gene expression profiling at CHI’s recent X-GEN Congress and Expotalks in San Diego.

Gene Expression Modulations

Her team was interested in looking at the pattern and gene expression modulations in lineage-specific differentiation of CD34+ cells in vitro, noted Dr. Raghavachari, further explaining that this approach circumvented problems with limited amounts of RNA that could be extracted from blood samples, as well as insufficiency in coverage of alternative splicing events.

“Whole transcriptome and exome analysis on both microarrays and sequencing platforms to identify genes that are significantly up- or downregulated during cell differentiation allowed us to discover novel alternatively spliced transcripts,” she further explained.

This gene expression approach helped her group determine which transcripts possessed functional roles in the regulation of cell proliferation, cell cycle signaling, and immune system development.

Baseed in Barcelona, Tamara Maes, Ph.D., chief scientific officer at Oryzon Genomics, also used gene expression profiling in answering specific protein activities in a myriad of organisms.

“We had no intention of becoming a microarray technology developer; we just had biological questions to answer and required reliable data, as well as flexibility to address the divergent needs of our clients,” she said.

“We incorporated a prototype mirror-based benchtop DNA microarray synthesizer and analyzer and developed proprietary DNA oligo design and data analysis programs that allowed us to design arrays for any species, and that included extensive features to monitor the quality of each experiment that we lacked in commercial arrays. As soon as we became confident enough in our own expertise, we selected Agilent as our provider,” Dr. Maes said.

Oryzon Genomics has been using gene expression profiling to study specific protein activities in a myriad of organisms.

Works with Variety of Organisms

Oryzon Genomics has been active in analyzing gene expression levels, splice variants, and copy number variations in a wide array of organisms, ranging from yeast and rice, to rodents. These efforts also allowed Dr. Maes’ group to develop their flagship drug discovery program on cancer and neurodegenerative disease.

According to Dr. Maes, the only method to ascertain Alzheimer’s disease was through the use of post-mortem histological analysis, with gene expression often considered as the major obstacle because of time-sensitive capturing of gene expression patterns during autopsy.

At the same time, comparative analysis of brains of Alzheimer’s disease patients and controls still presented subtle differences, thus further increasing their need to identify a more sensitive and reliable approach for gene expression profiling. In the case of cancer, “Our prototype technology allowed us to identify essential regulators of leukemia stem cell potential, possibly for the treatment of acute myeloid leukemia (AML), especially of genetically defined subtypes of AMLs carrying an MLL translocation,” Dr. Maes added.

In the department of genetics at the Washington University School of Medicine, Anne M. Bowcock, Ph.D., professor, applied deep sequencing techniques in identifying major modifications in the changes in mRNA expression profiles of psoriasis patients. Instead of concentrating on exons and mRNA for analysis, Dr. Bowcock screened miRNAs for changes in gene expression, including modifications in miRNA editing.

The advantage of using miRNAs in gene expression profiling in specific diseases involves the capacity to identify noncoding genomic sequences, which have lately been implicated in disease progression and diagnostics. In the case of skin differentiation, miRNA gene expression analysis could “reveal dramatic changes in global microRNA expression, reflecting defects in keratinocytes, immune cells, and vasculature,” according to Dr. Bowcock.

Her research program on temporal regulation of miRNAs was conducted using digital gene expression analysis, allowing her team to perform a global screen of 67 human skin samples, resulting in the detection of close to 100 novel miRNA molecules that were differentially expressed in psoriatic skin. Her group ensured that the digital read counts were highly reproducible and statistically robust.

In addition, results validation was also performed using quantitative RT-PCR on the same skin samples, further strengthening the signature phenotype of diseased skin and elucidating the expression of marker proteins in skin lesions.

Ultimate High-Throughput System

Gene expression profiling using exon array and next-gen sequencing may currently stand as the ultimate high-throughput system for monitoring protein profiles in specific diseases, yet other research groups have focused on further improving specific details of this method.

During the X-GEN meeting, Cliff Han, Ph.D., team leader, R&D, biosciences division, Los Alamos National Laboratory. He says it has the ability to classify DNA sequences generated from massive sequencing technologies.

This laptop-friendly, super-speed software may significantly improve read outs and sequence analysis of genes of interest, according to Dr. Han.

“This tool was actually designed to find out who’s there and what it’s actually doing,” said Joel Berendzen, Ph.D., one of the developers of Sequedex. He added that the concepts of phylogeny and functional analysis were integrated into a computational application that could tackle metagenome analysis at a higher level.

The software also considers the impact of the external environment on whole communities of organisms, identifying pieces of a protein that have been kept intact through natural selection using 10-mer signature amino acids reads.

“We want to know which genes are shared across the microbial kingdom, possibly later applying the same technology to other multicellular organisms in the tree of life,” said Nick Hengartner, Ph.D., a co-developer of Sequedex.

For now, interesting profiles have been generated from bacterial isolates from oral cavities, zeroing in on three genes that are possibly turned on or off in the presence of tooth decay.

“This is not a story of good or bad bacteria, but a story of three genes being transferred across the prokaryotic kingdom,” explained Dr. Berendzen.

In the near future, the group intends to look into differential expression profiles in microbes in an environment afflicted by oil spills, identifying what bacterial functions were activated amid the toxic compounds polluting the water, thus facilitating their survival.

Difference in gene expression in the oral transcriptome of 22 individuals with and 17 individuals without caries (tooth decay): All differences shown are > +/- 3 sigmas, with the largest peaks greater than 40 sigmas in both directions. The group with caries (blue peaks) had increased levels of just a few categories of genes, sugar utilization, group II introns, and a gene with a function unclassified by the scheme. The group without caries (red peaks) had vastly greater levels of nucleic acid conversions. No vertical streaks that would indicate increased levels of a particular phylogenetic group are observed. The data is consistent with a picture of a community that has chosen to get more energy from sugars rather than from saliva components and is using horizontal gene transfer to do so. Data from NCBI Short Read Archive study SRP007831. [Los Alamos National Laboratory]

Magnetic Particles

Another fine-tuned approach to gene expression profiling involves single-cell capture using magnetic particles.

“We simply pluck out one cell from blood and get a lot of information from this single unit,” said Abizar Lakdawalla, Ph.D., associate director of novel DNA sequencing technologies at Illumina.

“Magnetic capture allows us to determine the type of cell that is circulating in the blood, determine its gene expression profile, and identify its origin or tissue source,” continued Dr. Lakdawalla.

His group envisions that this technology can maximize the use of circulating tumor cells for early diagnosis of cancer, as well as to monitor the response of a patient to a specific therapeutic regimen. Coupling the magnetic capture technology with RNA-seq methodologies has generated information on gene fusions, SNPs, and transcript variants at the highest resolution possible—a single cell.

“The component of preparing RNA from a single cell may also be applied in embryogenesis, to resolve how one cell can change into a multicellular organism during development,” noted Dr. Lakdawalla.

Previous articleA First: Genome Sequencing Controls MRSA Outbreak
Next articleJanssen Launches Clinical Trials Database