A method commonly employed to study replicative ageing in yeast is laborious and slow. The use of miniaturized culture chambers opens the door for automated molecular analyses of individual cells during ageing.
Similarly to many cells in our body, the cells of budding yeast cannot replicate indefinitely. On division, a yeast cell gives rise to a mother cell and a 'fresh' daughter cell. The mother cell can produce, on average, only about 25 daughters before it dies. A test that measures the replicative lifespan of yeast cells has become a popular way to study ageing processes, and researchers have used it to identify genes and pathways that were later confirmed to have roles in longevity in animals1,2,3,4. However, such an assay is labour intensive and cannot be implemented in a high-throughput fashion5. Two studies, one by Lee et al.6 in Proceedings of the National Academy of Sciences and another by Xie et al.7 in Aging Cell, offer modified versions of the assay that are amenable to automation and that allow the study of ageing processes in yeast cells to be made in unprecedented detail. The techniques use tiny chambers to retain mother cells and wash away daughters, coupled to powerful microscopes capable of time-lapse photography.
In the conventional replicative ageing assay, the experimenter must look through a microscope and painstakingly remove each daughter cell after division using a small needle on the surface of thick, solid culture media (Fig. 1a). Moreover, just as with other organisms, there is significant variation in lifespan between individual yeast cells, even when they are genetically identical. This means that a minimum of 40 cells have to be interrogated to generate a reliable lifespan data set, which necessitates the manual removal of approximately 1,000 daughter cells.
Lee et al. and Xie et al. replaced the manual approach with transparent microfluidic devices that consisted of submillimetre-scale channels and tunnels through which nutrient broth flows in a controlled manner (Fig. 1b). Such a set-up allowed the authors to apply high-resolution microscopy techniques for tracking individual cells and molecular markers.
Subtle differences exist between the two systems, however. Lee and colleagues suspended the yeast cells between silicone micropads and thin cover glass. The micropads were slightly lifted by the hydrostatic pressure of the broth during loading of a cell suspension, and they held the mother cells after release of the pressure. Daughter cells were washed away because of their smaller size.
By contrast, Xie et al. trapped the cells in 'micro-jails' from which the daughters could escape through gates. The researchers also attached biotin molecules to the mothers' cell walls, causing these cells to adhere to the chambers' surfaces, which had been coated with avidin (a protein that binds biotin with high affinity). This ensured that only mother cells remained trapped, as the synthesis of new cell wall in yeast is confined to daughter cells, and no biotin was supplied after the initial labelling of the mother cells.
Interestingly, both groups of authors studied the same yeast strain but came up with contrasting results: the mean lifespan observed by Lee et al. was 25 cell divisions, whereas that found by Xie et al. was 18. A mean lifespan of approximately 27 was previously reported for the same strain when measured using the conventional replicative ageing assay8. Such discrepancies remain unexplained, but could be due to the different methods used. Importantly, however, the two microfluidic assays recapitulated the effects of known longevity mutants; for example, a strain lacking the gene FOB1 was found to have an increased lifespan, as expected.
Both studies documented significant diversity among individual cells of the same population as the cells aged and died. At the time of death, a spherical cell shape correlated with shorter lifespan, whereas an elongated shape was linked to longer lifespan. Furthermore, ageing and death were associated with profound changes in the morphology and function of intracellular organelles such as vacuoles (which became fragmented with age6) and mitochondria (which showed increased dysfunction7).
Xie and colleagues went one step further by demonstrating that they could track fluorescent molecular markers in individual ageing cells using high-resolution microscopy. In this way, they found that potentially harmful reactive oxygen species (ROS) increased in old cells. They also observed that expression of the Hsp104 chaperone — a protein that assists other proteins in folding — was inversely correlated with lifespan, although not with an accompanying increase in aggregates of damaged proteins as might have been anticipated.
The continuous imaging afforded by the two microfluidic approaches is likely to identify new and better molecular landmarks of cellular ageing, which are not accessible with the conventional replicative ageing assay. In addition, with further development, the microfluidic platforms will greatly facilitate high-throughput applications. Nevertheless, the conventional assay, albeit laborious, may remain the least biased of the three. For example, both microfluidic ageing assays rely on differences in size between mother and daughter cells. Because cell size seems to affect lifespan9, it needs to be established how the microfluidic assays perform in cases where the differences between mother and daughter may be diminished, or with mutants of altered cell size or shape. This might be less of a concern in Xie and colleagues' method, in which the labelling of the mothers with biotin helps confine them to the chamber, independently of their larger size.
It should also be noted that the microfluidic systems do not solve another shortcoming of the conventional assay: the poor yield of old cells, which are required if the researchers want to do biochemical tests on them. A different replicative ageing assay, the mother-enrichment programme10, may be better for exploring biochemical questions. In this system, yeast cells are genetically engineered to inhibit daughter-cell division in liquid culture, allowing the accumulation of large numbers of aged mother cells.
Clearly, by using the microfluidic systems and the mother-enrichment programme, it is becoming possible to combine biochemistry, cell biology and high-resolution lifespan measurement to make the most of the yeast replicative ageing model. Combining all these tools will enable scientists not only to identify molecular markers of ageing, but also to dissect the causal role of these factors in the ageing process. Furthermore, the anticipated increase in the throughput of the replicative ageing assays will usher in large-scale screening to identify small molecules that could modulate cellular ageing. With these tools in place, yeast replicative ageing is becoming an ideal experimental model that, when coupled to systems-biology approaches, may yield for the first time a holistic understanding of ageing in an organism.
Kaeberlein, M. Nature 464, 513–519 (2010).
Kaeberlein, M. et al. Science 310, 1193–1196 (2005).
Kaeberlein, M., McVey, M. & Guarente, L. Genes Dev. 13, 2570–2580 (1999).
Smith, E. D. et al. Genome Res. 18, 564–570 (2008).
Steffen, K. K., Kennedy, B. K., Kaeberlein, M. J. Vis. Exp. (28), e1209 (2009).
Lee, S. S., Vizcarra, I. A., Huberts, D. H. E. W., Lee, L. P. & Heinemann, M. Proc. Natl Acad. Sci. USA 109, 4916–4920 (2012).
Xie, Z. et al. Aging Cell http://dx.doi.org/10.1111/j.1474-9726.2012.00821.x (2012).
Kaeberlein, M., Kirkland, K. T., Fields, S. & Kennedy, B. K. PLoS Biol. 2, 1381–1387 (2004).
Yang, J. et al. Cell Cycle 10, 144–155 (2011).
Lindstrom, D. L. & Gottschling, D. E. Genetics 183, 413–422 (2009).