June 2009


Aqueous Phase Separation in Living Cells

Scientists have learned a great deal about cells and the biomaterials (DNA, proteins, metal ions, etc) within them. This knowledge has helped scientists treat diseases such as cholera, elucidate the genetics of famine and other health questions, clean up chemical waste, and synthesize anti-HIV drugs and other valuable materials, such as quantum dots.

However, there is still a long way to go, even on fundamental principles. One such important (yet unanswered) question, of biological and technological importance, is how the cell cytoplasm (the interior of cells) organizes itself into a complex network of distinct functional subunits.

The common view of cytoplasmic organization is that cells possess many interior compartments, each surrounded by a physical barrier (such as a phospholipid membrane). However, proteins are known to localize to specific regions of the cytoplasm, in the absense of physical barriers.

This phenomenon is known as protein microcompartmentation. What enables it, and consequently the cellular functions it enables (e.g., DNA repair and intracellular communication)? Numerous hypotheses have been put forward, some of them heavily based in the biological sciences, and others more grounded in the physical sciences.

The aqueous phase separation hypothesis.

One of these hypotheses is that protein microcompartmentation is enabled by aqueous phase separation, which is typically the spontaneous segregation of proteins into multiple water-based (aqueous) phases. Proponents of the aqueous phase separation hypothesis point out that the cell cytoplasm possesses a high concentration of proteins (up to tens of weight percent).

These are conditions that are known to enable aqueous phase separation of proteins (and other polymers) in test tubes. Thus, a physical feature of the cell cytoplasm (its high protein concentration) may drive spontaneous intracellular organization.

However, to date, aqueous phase separation has not been observed in living cells. Most relevant to biology, it has been observed in nonliving cells exposed to harsh conditions, and in artificial constructs designed to primitively mimic cells.

The observation of aqueous phase separation in living cells wouldn't necessarily demonstrate that it is a normal feature of the cell cytoplasm. Howver, such an observation would lend some weight to the aqueous phase separation hypothesis, and justify the search for evidence of the phenomenon as part of normal cellular physiology.

Carlos Filipe (McMaster University, Canada) and coworkers have induced aqueous phase separation of polypeptides (proteins) in genetically engineered bacteria and plant cells. After a cell had produced a certain amount of the desired protein, the proteins formed their own aqueous phase(s), and localized to one or both ends (poles) of the cell.

Genetically engineering the cells.

The scientists genetically engineered their cells to produce elastin-like polypeptides. Elastin is a protein found in certain cells that enable them to resume their typical shape after cellular deformation.

Elastin-like polypeptides are synthetic molecules that superficially resemble the protein elastin. Importantly for Filipe and coworkers, elastin-like polypeptides are known to undergo aqueous phase separation in response to appropriate stimuli.

The scientists' elastin-like polypeptides were affixed to a fluorescent protein as a part of the genetic engineering process. This enabled the scientists to localize the polypeptides, and prove the fluid nature of the protein phases, in living cells with fluorescence microscopy.

Confirming aqueous phase separation.

The scientists first needed to demonstrate that their polypeptide undergoes aqueous phase separation, both in microscopic droplets and in living cells. For both purposes, their primary technique was fluorescence recovery after photobleaching.

The basis of fluorescence recovery after photobleaching is to shine a high-power laser beam at a small region, irreversibly destroying fluorescence from molecules within the region. Afterwards, nonfluorescent (damaged) molecules move out of the region, and fluorescent (undamaged) molecules move into the region, resulting in fluorescence recovery within the small region.

The rate of fluorescence recovery is related to the mobility of molecules within the small region being monitored. For these scientists' purposes, they were interested in demonstrating that the molecules within the presumed aqueous phases were fluid, not immobile protein clumps.

They observed a large amount of fluorescence recovery in microscopic droplets of phase-separated elastin-like polypeptides, demonstrating the fluid nature of the aqueous phases. Complete recovery was not observed, due to the inherent mobility of the polypeptides, and some unavoidable destruction of the fluorescent protein affixed to the polypeptides.

Whether or not the elastin-like proteins underwent aqueous phase separation in living cells depended on how long the cells were given to synthesize the protein (i.e., protein concentration in the cell). Some aqueous phase separation was observed after 10 hours, more after 20 hours, and enough to coalesce into larger droplets after 36 hours; the scientists observed a large amount of fluorescence recovery in the cells, similarly to within the microscopic droplets.

Functional protein-enriched microcompartments.

The scientists used fluorescent dyes to confirm that the elastin-like polypeptide aqueous phases collected at one or both ends (poles) of the cell. Furthermore, nucleic acids (genetic material) and ribosomes (protein synthesis "machinery") were excluded from the polypeptide phases.

These observations are especially noteworthy. They demonstrate the spontaneous segregation of the polypeptides from other specialized regions of the cell.

From a biotechnology standpoint, this segregation may be useful for a number of reasons. These include preventing polypeptide aggregation by nonspecifically directing them to a region of especially high polypeptide concentration, in-cell synthesis and subsequent isolation of toxic proteins, and possibly for preventing degradation by other intracellular proteins such as proteases (tested here in bulk solution experiments, not within living cells).

From a synthetic biology standpoint, this segregation is useful for demonstrating the induction of protein microcompartmentation, via aqueous phase separation. No other technique currently developed is capable of generating these two phenomena in an unambiguous manner, in living cells.

Overall evaluation.

Filipe and coworkers have induced protein microcompartmentation, via aqueous phase separation of elastin-like polypeptides, in genetically engineered living cells. This development may enable scientists to test hypotheses of intracellular organization and assembly, in model systems that bridge the divide between entirely synthetic constructs and living cells in a physiological state.

for more information:
Ge, X., Conley, A. J., Brandle, J. E., Truant, R., & Filipe, C. D. M. (2009). In Vivo Formation of Protein Based Aqueous Microcompartments Journal of the American Chemical Society, 131 (25), 9094-9099 DOI: 10.1021/ja902890r