Overview Research Interests of the Braakman/Tabak/Rüdiger Lab

PROTEIN FOLDING & CHAPERONES

Proteins are the workhorses of the living cell. They may differ in sequence, shape and function, but have in common that they all have to fold into specific three-dimensional structures, which are mandatory for proper function. Protein structures however are not rigid. Instead, proteins have a dynamic life style, which may involve unfolding and refolding, complex association and dissociation. Stress, but also many physiological events, require proteins to surrender their structure or to regain it at a later stage. Understanding protein folding within the complex environment of the living cell is a key question in modern biology and biological chemistry.

The cell controls many of these protein folding processes itself, others are forced onto the cell from the environment. Essential processes are events such as de novo synthesis of proteins, protein translocation into different compartments, or control of the activity of regulatory proteins. A major force from outside that damages protein structures is heat stress due to increase in temperature. Similar problems can be caused by chemical compounds such as solvents or heavy metals. DNA damage may lead indirectly to protein folding problems because the mutated proteins are often less stable than the wild type. Protein folding defects have important medical implications. They are associated e. g. with amyloid diseases such as Alzheimer and prion diseases such as Creutzfeldt-Jakob-Disease, but they are also a major cause of cancer due to destabilising mutations in the tumour suppressor protein p53. Likewise, several genetic disorders relate to protein folding defects: mutations in the CFTR protein that lead to its misfolding cause cystic fibrosis, while folding deficient LDL receptor protein variants cause familial hypercholesterolemia.

Every cell in every organism owns an arsenal of molecular chaperones to control folding and unfolding of proteins, or to react on protein unfolding during stress conditions. Most molecular chaperones are members of evolutionary conserved families: Hsp100, Hsp90, Hsp70 and their DnaJ (Hsp40) co-chaperones, the chaperonins Hsp60, and the small heat shock proteins (sHsp). Nearly all organisms have at least one homologue of each of these classes. Apart from chaperones also folding enzymes, such as peptidyl-prolyl cis-trans isomerases, assist protein folding in the cell. In contrast to prokaryotes, eukaryotes contain various organelles separated by lipid membranes. Translocation of fully folded proteins across membranes is often avoided. Instead, folding occurs in different compartments of the eukaryotic cell. The cytosol, mitochondria, chloroplasts in plants, and the Endoplasmic Reticulum (ER), each have their own ensemble of chaperones and folding enzymes. In addition, some chaperones exist that are exclusive to specific cell compartments or to bacteria.

CHAPERONE-SUBSTRATE INTERACTIONS (Stefan Rüdiger)

Our aim is to understand how protein stability influences cellular processes, and to analyse how the cell is able to buffer eventual damage to proteins. Severely destabilised proteins are usually degraded. Interesting is what happens to proteins that are damaged in a way that they can be kept in solution by molecular chaperons. Chaperones act as the cellular thermometer: the binding specificity of a chaperone provides the cellular definition of what is an unfolded protein. Studying the structural properties of chaperone substrates should give a direct insight about the structural properties of unfolded proteins inside the cell under physiological condition.

In the past we have established what makes an unfolded protein a substrate of DnaK as an example of the Hsp70 family, and for a few more chaperones. The Hsp70s seem to serve as versatile 'generic’ chaperones, many of which are involved in various folding processes. The most abundant chaperone in the eukaryotic cytosol, however, is Hsp90. It seem peculiar that the specificity of Hsp90 chaperones seems to be much more narrow that of their Hsp70 colleagues. Substrates of Hsp90 are mainly kinases and transcription factors, most of them are oncogenes.

We characterised previously the structural properties of an oncogenic Hsp90 substrate, the tumour suppressor p53, in the Hsp90-bound state and found it to be unfolded. The destabilisation of p53 by structural mutations is crucial for the shutdown of apoptosis and DNA repair pathways in tumour cells. It seems likely that destabilisation of proteins could affect other central processes in the cell as well. It is, however, an open question whether substrates of Hsp90 are all unfolded, or whether p53 is a special example. The activity of Hsp90 is controlled by an ATP cycle, which is modulated by interaction with other chaperones and co-factors. However, it is so far unknown what happens structurally to the substrate during the cycle and to which extent Hsp90 is involved in an active folding process.

• THE HSP90 CHAPERONE CYCLE
  Hsp90 binds unfolded substrate protein, followed by binding of ATP. Hydrolysis of ATP is coupled to substrate release, and dissociation of ADP allows to start the cycle again. Co-factors control steps within the cycle. The molecular mechanisms of substrate recognition and substrate folding are poorly understood (Hsp90 is shown in red, the substrate as black/yellow circle).

We use biochemical and biophysical techniques to find out features that characterise proteins as chaperone substrates, which can be seen as the cellular definition of unfolded protein, in particular of substrates of Hsp90 chaperones. We aim at elucidating the consequences of protein instability on their cellular function. Related to this is the question to which extent tolerated instability is an essential regulatory feature. Many of the substrates are bound to Hsp90 until they are activated, e. g. by binding a co-factor such as a steroid. p53 is unfolded when bound to Hsp90 - are all regulatory proteins that are Hsp90 substrates present in the cell in an unfolded state until they are activated? And when they are activated, how do they fold? We would like to establish what features regulatory proteins have in common that in the cell these but not other proteins are recognised by Hsp90: how does Hsp90 select its substrates? We want to characterise the structural properties of Hsp90 substrates: do substrate proteins unfold upon Hsp90 binding, or do they adopt a specific structure? And finally, what is the molecular mechanism of Hsp90 chaperone activity?

PROTEIN FOLDING IN THE ER (Ineke Braakman)

The ER is the first compartment of the secretory pathway. It is the cradle of all cell surface proteins, proteins that get secreted and in addition proteins that reside in any compartment along the exocytic and endocytic pathways. N-terminal hydrophobic signal peptides destine these proteins to enter the ER lumen, where they fold and oligomerize with assistance of ER resident chaperones and folding enzymes. The association of incompletely folded or misfolded ER client proteins to the ER resident folding factors inherently leads to their retention. Conversely, fully folded and oligomerized ER clients are released from the chaperones. They exit the ER to travel to their final destination in or outside the cell.

To chaperone hydrophobic interactions, the ER harbors folding factors that have equivalents in other folding compartments: members of the HSP90 family (GRP94), of the HSP70 family (BiP and GRP170) and several J-domain containing proteins. The ER also contains a number of PPIases. Typical for most ER clients is that their signal peptides are cleaved off. Special to the ER is also that N-linked glycans are added to newly synthesized ER clients. N-glycans are important for the folding process. Most importantly, they facilitate that the lectins calnexin and calreticulin can associate with folding intermediates to promote their maturation by a mechanism unique to the ER.

• SCHEMATIC REPRESENTATION OF PROTEIN FOLDING IN THE ER
  ER client proteins are co-translationally translocated into the ER lumen, where often their N-terminal signal peptide (green) is cleaved off and where they obtain N-linked glycans (yellow). Both co-translational and post-translational folding events are chaperoned by an array of ER resident folding factors (light blue). Disulfide bond (red) formation is assisted by PDI or its family members. Oxidative folding proceeds until monomeric subunits are fully folded. Only when all subunits of ER-client proteins have folded correctly and have assembled into oligomers, chaperones and folding enzymes release them. ER clients subsequently exit from the ER.

Arguably the most distinctive element of ER folding is that it is accompanied by disulfide bond formation. In contrast to the co-translational signal peptide removal and N-glycosylation, thiol-oxidation can be a long lasting process. Oxidative folding is dependent on a collection of proteins that are unique to the ER. Thiol-oxidoreductases of the protein disulfide isomerase (PDI) family catalyze the formation of disulfide bonds. Oxidized PDI contains disulfide bonded CXXC motifs, which act as donor of oxidizing equivalents. These are transferred onto maturing secretory proteins via formation of mixed disulfides. In the process PDI is reduced. Thus, to sustain the net flux of disulfides into proteins that get secreted, PDI must be constantly reoxidized; a responsibility of the Ero1 protein. A second important task for PDI is the isomerization of disulfide bonds. In that way, aberrant disulfide bonds are disentangled in favor of the formation of native disulfide bonds. Illustrative for its dedication to disulfide bond formation, the ER harbors a growing list of PDI-like proteins.

To better understand protein folding in the ER, we focus on a set of four model proteins. They represent the complete spectre of proteins passing through the ER: Influenza virus HA (of average complexity), HIV Envelope glycoprotein (rich in N-linked glycans), the low density lipoprotein receptor (rich in disulfide bonds) and the cystic fibrosis related CFTR protein (12 transmembrane domains). These proteins are subjected to thorough conformational studies in which antibody-reactivity and the formation of disulfide bonds are followed. Combined with mutant analysis, we now have an extensive picture of the folding process for each protein in the ER. In addition, we characterize known and newly identified (putative) ER chaperones and folding enzymes for their role in the biosynthesis of our model proteins. To find candidate ER chaperones, we exploit ER expansion during B cell differentiation or during ER stress.

ER-PEROXISOME CONNECTION (Henk Tabak)

Apart from its important role in protein secretion the ER is also the factory that gives rise to a number of distinct intracellular compartments such as Golgi, endosomes, lysosomes, secretory granules and cell membrane. Our recent work has shown that peroxisomes share kinship with this family of organelles. Some peroxisomal membrane proteins arrive first in the ER membrane and take the lead in a developmental process that recruits membrane from the ER, additional proteins to build up the peroxisome-specific protein import machinery, finally resulting in the formation of fully functional and metabolically active organelles.

Communication between the secretory and endocytic pathways involve SNAREs, SNAPs, SNFs, Rabs etc. An interesting question to answer will be to investigate if the new peroxisomal offshoot of this compartment makes use of similar parts or that entirely new proteins developed to support the various steps involved in the maturation pathway outlined above. Moreover, this new insight puts the traditional view that peroxisomes, together with mitochondria and chloroplasts, arose as endosymbionts in the developing eukaryotic cell into a totally new perspective. The rapidly expanding field of comparative genomics may help to bring new facts to light on the evolutionary development of peroxisomes: an old or a more recent invention of nature?

home Department of Cellular Protein Chemistry