The ribosome regulates flavodoxin folding

During and after their translation by the ribosome, folding of polypeptides to biologically active proteins is of vital importance for all living organisms. Gaining knowledge about nascent chain folding is required to enhance our understanding of protein folding in the cell. This in turn allows us to obtain insights into factors responsible for protein misfolding, aggregation, and, potentially, for numerous devastating pathologies.

In Chapter 1 the model protein flavodoxin is introduced. Also theories about protein folding are presented, which led to the concept of the “folding energy landscape”. Flavodoxin folds via a misfolded off-pathway intermediate, which is molten globular and forms extensively during its refolding in vitro.

In Chapter 2 we show that flavodoxin also populates an on-pathway molten globule during its folding. In the F44Y variant of apoflavodoxin, lowering the ionic strength induces the off-pathway molten globule state. By adding the cofactor FMN, we could follow aspects of the folding of this protein, as off-pathway molten globular flavodoxin first has to unfold and subsequently refold before FMN can bind. Thus, presence of the off-pathway molten globule retards FMN binding. We determined the presence of the off-pathway molten globule at decreasing ionic strengths with cofactor binding kinetics and polarized time-resolved tryptophan fluorescence spectroscopy. Comparison of both data sets revealed the presence of another, concurrently present molten globule. This species is most likely on-pathway to native protein. To our knowledge this is the first time that two concurrent molten globules have been discovered that reside on folding routes of decidedly different nature (i.e., on- and off-pathway ones).

While much work has been done on the folding of flavodoxin in vitro, the next step is to elucidate how this protein folds in vivo. In Chapter 3 the first insights into cotranslational flavodoxin folding are presented. By using ribosomal nascent chains (RNCs) we could determine that when flavodoxin is fully exposed outside the ribosome it can bind its cofactor. However, while its five C-terminal amino acids are still sequestered in the ribosomal exit tunnel, the protein is in a non-native state and cannot bind FMN. Thus the last step in production of this flavoprotein in vivo is the binding of cofactor.

Chapter 4 reveals the influence of the ribosome on formation of the off-pathway molten globule of flavodoxin. By using RNCs of the F44Y variant of apoflavodoxin, we proved that the ribosome restrains formation of this molten globule. This discovery was possible by exploiting the findings of Chapter 2 and Chapter 3, namely that cofactor binding kinetics slow down when off-pathway molten globule is present and that a fully exposed, natively folded flavodoxin nascent chain binds FMN. For F44Y RNCs no retardation in FMN binding occurs, whereas cofactor binding slows down in case of isolated, full-length F44Y in the molten globule state. Thus the ribosome restrains formation of molten globules in stalled nascent flavodoxin. This is possibly due to electrostatic repulsion of the nascent chain by the ribosomal surface, as both are negatively charged, leading to entropic stabilization of native protein at physiological ionic strength.

In Chapter 5 we review experiments and simulations concerning the folding of flavodoxins and CheY-like proteins, which share the flavodoxin-like fold. These proteins form intermediates that are off-pathway to native protein and several of these species are molten globules. This susceptibility to frustration is caused by the more rapid formation of an α-helix compared to a β-sheet, particularly when a parallel β-sheet is involved. The experimentally characterized off-pathway intermediates seem to be of α-helical nature. We discuss the probing of the cotranslational folding of flavodoxin as a first step towards a molecular description of how flavodoxin-like proteins fold in vivo.

Finally, Chapter 6 touches upon the implications of our findings and possible applications in biotechnology, health and disease. A finding that has potential application is the role FMN has as a chemical chaperone. This chemical chaperone can already work at the cotranslational level, as binding of FMN stabilizes a nascent chain and thereby protects the nascent chain against degradation.