Protein Folding Inside the Cell

Through pioneering work during the last three decades, using, for example, protein engineering (1) and energy landscape theory (2), significant progress has been made to pinpoint mechanisms and driving forces important for protein folding. We now know mechanistic details of the folding reactions of many proteins. In general, polypeptide folding is viewed as a random search of conformational space on a more (resulting in populated intermediates) or less (resulting in two-state reaction) rugged funneled-shaped energy surface. However, in reality, proteins fold inside cells that are environments very different from that of a dilute buffer solution most often used in in vitro experiments. The cell compartments (cytoplasm, endoplasmic reticulum (ER), nucleus, etc.) are full of other proteins, membranes, and DNA; the level and heterogeneity of biomolecules may vary depending on the compartment. It is estimated that up to 40% of the available volume in a cell is occupied by other biomolecules (3). The crowded environment results in increased viscosity, excluded volume effects, and the amplified opportunity for specific as well as nonspecific intermolecular interactions. These environmental factors are not accounted for in the fundamental studies of protein folding mechanisms executed during the last decades. The question thus arises how these effects—present when polypeptides normally fold in vivo—modulate protein folding reactions. To obtain a complete understanding of protein folding in vivo, we need to investigate the role and magnitude of these effects.

In a study reported in this issue (page 421), Gruebele et al. perform original experiments to begin to address the above question. They have used fluorescence imaging methods to study stability and folding dynamics of a protein inside three different compartments of bone tissue cells. The design of the experiments is clever. First, they use a destabilized version of the protein phosphoglycerate kinase (PKG) that results in selective unfolding of this protein upon a temperature jump of only a few degrees. Second, the PKG polypeptide is connected in each end to another (more stable) protein, green fluorescent protein and red fluorescent protein, which makes possible donor (green) to acceptor (red) Förster resonance energy transfer (FRET) measurements. In this arrangement, the FRET efficiency reports on the compactness of PKG and can be used as a probe of PKG unfolding. Third, they attached localization tags that direct the PKG-FRET construct to either the ER or the nucleus; in the absence of tag, PKG accumulates in the cytoplasm. Fourth, using a home-built fast relaxation imaging microscope, kinetic and equilibrium fluorescence signals from individual cells could be measured upon infrared laser irradiation that triggers temperature changes.

Gruebele et al. find that both unfolded and folded states of PKG are more compact inside the three cell compartments as compared to these states when in dilute buffer in vitro. In agreement, excluded volume theory suggests a compaction of the unfolded state in crowded conditions. Macromolecular crowding effects on the folded structure in vitro have been reported for a few other proteins (e.g.,(4)). It is tempting to speculate that environment-induced shape changes in folded proteins may be a way to modulate activity in vivo. Interestingly, it is found that the PKG folding speed is faster in the nucleus than in the ER and the cytoplasm despite the same level of crowding. In buffer in vitro, folding of PKG is a complex reaction involving multiple steps. This kinetic mechanism remains for PKG when probed in the cytoplasm and the nucleus, but the folding kinetics detected for PKG in the ER appears more two-state like. This implies that cellular environments can tune the folding landscape of a protein so that it becomes smoother than in vitro. This phenomenon was recently reported for another protein as a direct result of excluded volume effects in vitro (5).

Taken together, the Gruebele et al. work reveals notable differences between folding in vitro and in vivo. However, the most exciting conclusion is the fact that there are differences in folding behavior depending on the in vivo compartment. This indicates that it is of utmost importance to include cell-compartment-specific chemical attributes of crowders when performing in silico computations to mimic in vivo protein reactions. It is clear from this study that excluded volume effects, although prominent, are not the sole determinant of differences in in vitro and in vivo results. To sort out the environmental contributions affecting protein folding and function in vivo on a molecular level, in vitro biophysical experiments need to be designed that account for effects of excluded volume, viscosity, and nonspecific/specific interactions in controlled ways.

There are two limitations to note in this study. First, the temperature-induced unfolding reaction involves changes in tertiary interactions—PKG secondary structure remains intact during the temperature jumps. Thus, the conclusions regarding changes in folding speed and mechanism concern only tertiary contacts. It will be important to design future experiments that involve complete protein unfolding to probe the whole energy landscape. Second, the construct used here is a three-protein construct: thus it is large and likely not spherical. In vitro and in silico experiments have shown that excluded volume effects depend on protein/crowder ratio as well as protein shape. It is possible that the in vivo effects observed are exaggerated because the construct is larger than PKG alone.

The question of how proteins fold in vivo is an emerging topic, and future studies—using a combination of in silico, in vitro, and in vivo approaches—are essential to obtain deeper insight. Not only is this topic important to identify the basic principles of protein folding in vivo, it is also relevant for a better comprehension of protein misfolding diseases and conformational changes linked to enzyme turnover.