The therapy, known as adoptive T cell transfer, has shown encouraging results in clinical trials. This treatment involves removing disease-fighting immune cells called T cells from a cancer patient, multiplying them in the laboratory and then infusing them back into the patient's body to attack the cancer. The effectiveness of this therapy, however, is limited by the finite lifespan of T cells -- after many divisions, these cells become unresponsive and inactive.

Researchers at Georgia Tech and Emory University have addressed this limitation by developing a microfluidic device for sample handling that allows a statistical model to be generated to evaluate cell responsiveness and accurately predict cell "age" and quality. Being able to assess the age and responsiveness of T cells -- and therefore transfer only young functional cells back into a cancer patient's body -- offers the potential to improve the therapeutic outcome of several cancers.

"The statistical model, enabled by the data generated with the microfluidic device, revealed an optimal combination of extracellular and intracellular proteins that accurately predict T cell age," said Melissa Kemp, an assistant professor in the Wallace H. Coulter Department of Biomedical Engineering at Georgia Tech and Emory University. "Knowing this information will help facilitate the clinical development of appropriate T cell expansion and selection protocols."

Details on the microfluidic device and statistical model were published in the March issue of the journal Molecular & Cellular Proteomics. This work was supported by the National Institutes of Health, Georgia Cancer Coalition, and Georgia Tech Integrative Biosystems Institute.

Currently, clinicians measure T cell age by using multiple assays that rely on measurements from large cell populations. The measurements determine if cells are exhibiting functions known to appear at different stages in the life cycle of a T cell.

"Since no one measurement is a perfect predictor, it is advantageous to concurrently sample multiple proteins at different time points, which we can do with our microfluidic device," explained Kemp, who is also a Georgia Cancer Coalition Distinguished Professor. "The wealth of information we get from our device for a small number of cells far exceeds a single measurement from a population the same size by another assay type."

For their study, Kemp, electrical engineering graduate student Catherine Rivet and biomedical engineering undergraduate student Abby Hill analyzed CD8+ T cells from healthy blood donors. They acquired information from 25 static biomarkers and 48 dynamic signaling measurements and found a combination of phenotypic markers and protein signaling dynamics -- including Lck, ERK, CD28 and CD27 -- to be the most useful in predicting cellular age.

To obtain biomarker and dynamic signaling event measurements, the researchers ran the donor T cells through a microfluidic device designed in collaboration with Hang Lu, an associate professor in the Georgia Tech School of Chemical & Biomolecular Engineering. After stimulating the cells, the device divided them into different channels corresponding to eight different time points, ranging from 30 seconds to seven minutes. Then they were divided again into populations that were chemically treated to halt the biochemical reactions at snapshots in time to build up a picture of the signaling events that occurred as the T cells responded to antigen.

"While donor-to-donor variability is a confounding factor in these types of experiments, the technological platform minimized the experimental data variance and allowed stimulation time to be precisely controlled," said Lu.

With the donor T cell data, the researchers developed a model to assess which biomarkers or dynamical signaling events best predicted the quality of T cell function. The model found the most informative data in predicting cellular age to be the initial changes in signaling dynamics.

"Although a combination of biomarker and dynamic signaling data provided the optimal model, our results suggest that signaling information alone can predict cellular age almost as well as the entire dataset," noted Kemp.

In the future, Kemp plans to use this approach of combining multiple cell-based experiments on a microfluidic chip to integrate single-cell information with population-averaged techniques, such as multiplexed immunoassays or mass spectrometry.

This project is supported in part by the National Institutes of Health (NIH)(Grant No. R21CA134299). The content is solely the responsibility of the principal investigator and does not necessarily represent the official views of the NIH.

Research News & Publications Office
Georgia Institute of Technology
75 Fifth Street, N.W., Suite 314
Atlanta, Georgia 30308 USA

Media Relations Contacts: Abby Robinson (abby@innovate.gatech.edu; 404-385-3364) or John Toon (jtoon@gatech.edu; 404-894-6986)

Writer: Abby Robinson

The team of researchers from the Georgia Institute of Technology, Michigan State University and the Pacific Northwest National Laboratory analyzed the gene sequences, proteins expressed and physiology of 10 strains of Shewanella. They believe the study results will help researchers choose the best Shewanella strain for bioremediation projects based on each site's environmental conditions and contaminants.

The findings, which further advance the understanding of the enormous microbial biodiversity that exists on the planet, appear in the early online issue of the journal Proceedings of the National Academy of Sciences. This research was supported by the U.S. Department of Energy through the Shewanella Federation consortium and the Proteomics Application project.

Similar to a human breathing in oxygen and exhaling carbon dioxide, many Shewanella microbes have the ability to "inhale" certain metals and compounds and convert them to an altered state, which is typically much less toxic. This ability makes Shewanella very important for the environment and bioremediation, but selecting the best strain for a particular project has been a challenge.

"If you look at different strains of Shewanella under a microscope or you look at their ribosomal genes, which are routinely used to identify newly isolated strains of bacteria, they look identical. Thus, traditional microbiological approaches would suggest that the physiology and phenotype of these Shewanella bacteria are very similar, if not identical, but that is not true," explained Kostas Konstantinidis, an assistant professor in the Georgia Tech School of Civil and Environmental Engineering. Konstantinidis, who also holds a joint appointment in the Georgia Tech School of Biology, led the research team in analyzing the data.

Using the traditional method for determining interrelatedness between microbial strains -- sequencing of the 16S ribosomal gene -- the researchers determined that the 10 strains belonged to the same genus. However, the technique was unable to distinguish between most of the strains or define general properties that would allow the researchers to differentiate one strain from another. To do that, they turned to genomic and whole-cell proteomic data.

By comparing the 10 Shewanella genomes, which were sequenced at the Department of Energy's Joint Genome Institute, the research team found that while some of the strains shared 98 percent of the same genes, other strains only shared 70 percent. Out of the almost 10,000 protein-coding genes in the 10 strains, nearly half -- 48 percent -- of the genes were strain-specific, and the differences in expressed proteins were consistently larger than their differences at the gene content level.

"These findings suggest that similarity in gene regulation and expression constitutes an important factor for determining phenotypic similarity or dissimilarity among the very closely related Shewanella genomes," noted Konstantinidis. "They also indicate that it might be time to start replacing the traditional microbiology approaches for identifying and classifying new species with genomics- or proteomics-based methods."

Upon further analysis, the researchers found that the genetic differences between strains frequently reflected environmental or ecological adaptation and specialization, which had also substantially altered the global metabolic and regulatory networks in some of the strains. The Shewanella organisms in the study appeared to gain most of their new functions by acquiring groups of genes as mobile genetic islands, selecting islands carrying ecologically important genes and losing ecologically unimportant genes.

The most rapidly changing individual functions in the Shewanellae were related to "breathing" metals and sensing mechanisms, which represent the first line of adaptive response to different environmental conditions. Shewanella bacteria live in environments that range from deep subsurface sandstone to marine sediment and from freshwater to saltwater. All but one of the strains was able to reduce several metals and metalloids. That one exception had undertaken a unique evolution resulting in an inability to exploit strictly anaerobic habitats.

"Let's say you have a strain of Shewanella that is unable to convert uranium dissolved in contaminated groundwater to a form incapable of dissolving in water," explained Konstantinidis. "If you put that strain in an environment that contains high concentrations of uranium, that microbe is likely to acquire the genes that accept uranium from a nearby strain, in turn preventing uranium from spreading as the groundwater flows."

This adaptability of bacteria is remarkable, but requires further study in the bioremediation arena, since it frequently underlies the emergence of new bacterial strains. Konstantinidis' team at Georgia Tech is currently investigating communities of these Shewanella strains in their natural environments to advance understanding of the influence of the environment on the evolution of the bacterial genome and identify the key genes in the genome that respond to specific environmental stimuli or conditions, such as the presence of heavy metals.

Ongoing studies should broaden the researchers' understanding of the relationship between genotype, phenotype, environment and evolution, he said.

Research News & Publications Office
Georgia Institute of Technology
75 Fifth Street, N.W., Suite 100
Atlanta, Georgia 30308 USA

Media Relations Contacts: Abby Vogel (404-385-3364); E-mail: (avogel@gatech.edu) or John Toon (404-894-6986); E-mail: (jtoon@gatech.edu).

Writer: Abby Vogel

Many researchers begin modeling biological systems with simple organisms, such as bacteria, that are easy to culture, manipulate genetically, maintain under controlled conditions and examine in the laboratory. Such microorganisms hint at functionality in larger organisms yet are less complex. Georgia Tech biomedical engineer Eberhard Voit, in collaboration with Helena Santos, a professor at the Universidade Nova de Lisboa in Portugal, showed that high-precision, dynamic experimental data can be combined with nonlinear mathematical modeling to characterize mechanisms in the bacterium Lactococcus lactis.

To view full article, visit: http://www.livescience.com/animals/090605-sl-voit.html

For more information on Voit lab, visit: Eberhard Voit