Umeå University's logo

The transcriptional activator PrfA, a member of the Crp/Fnr family, controls the expression of some key virulence factors necessary for infection by the human bacterial pathogen Listeria monocytogenes. Phenotypic screening identified ring-fused 2-pyridone molecules that at low micromolar concentrations attenuate L. monocytogenes infectivity by reducing the expression of virulence genes, without compromising bacterial growth. These inhibitors bind the transcriptional regulator PrfA and decrease its affinity for the consensus DNA binding site. Structural characterization of this interaction revealed that one of the ring-fused 2-pyridones, compound 1, binds within a hydrophobic pocket, located between the C- and N-terminal domains of PrfA, and interacts with residues important for PrfA activation. This indicates that these inhibitors maintain the DNA-binding helix-turn-helix motif of PrfA in a disordered state, thereby preventing a PrfA:DNA interaction. Ring-fused 2-pyridones represent a new class of chemical probes for studying virulence in L. monocytogenes.

Copper (Cu) is an essential trace element but toxic in free form. After cell uptake, Cu is transferred, via direct protein-protein interactions, from the chaperone Atox1 to the Wilson disease protein (WD) for incorporation into Cu-dependent enzymes. Cu binds to a conserved C1XXC2 motif in the chaperone as well as in each of the cytoplasmic metal-binding domains of WD. Here, we dissect mechanism and thermodynamics of Cu transfer from Atox1 to the fourth metal binding domain of WD. Using chromatography and calorimetry together with single Cys-to-Ala variants, we demonstrate that Cu-dependent protein heterocomplexes require the presence of C-1 but not C-2. Comparison of thermodynamic parameters for mutant versus wild type reactions reveals that the wild type reaction involves strong entropy-enthalpy compensation. This property is explained by a dynamic inter-conversion of Cu-Cys coordinations in the wild type ensemble and may provide functional advantage by protecting against Cu mis-ligation and bypassing enthalpic traps.

After Ctr1-mediated cell uptake, copper (Cu) is transported by the cytoplasmic Cu chaperone Atox1 to P_1B type ATPases ATP7A and ATP7B in the Golgi network, for incorporation into Cudependent enzymes. Atox1 is a small 68-residue protein that binds Cu in a conserved CXXC motif; it delivers Cu to target domains in ATP7A/B via direct protein-protein interactions. Specific transcription factors regulating expression of the human Cu transport proteins have not been reported although Atox1 was recently suggested to have dual functionality such that it, in addition to its cytoplasmic chaperone function, acts as a transcription factor in the nucleus. To examine this hypothesis, here we investigated the localization of Atox1 in HeLa cells using fluorescence imaging in combination with in vitro binding experiments to fluorescently labeled DNA duplexes harboring the proposed promotor sequence. We found that whereas Atox1 is present in the nucleus in HeLa cells, it does not bind to DNA in vitro. It appears that Atox1 mediates transcriptional regulation via additional (unknown) proteins.

Many processes in living systems occur through transient interactions among proteins. Those interactions are often weak and are driven by small changes in free energy. Due to the short-living nature of these interactions, our knowledge about driving forces, dynamics and structures of these types of protein-protein heterocomplexes are though limited. This is especially important for cellular copper (Cu) trafficking:

Copper ions are essential for all eukaryotes and most bacteria. As a cofactor in many enzymes, copper is especially vital in respiration or detoxification. Since the same features that make copper useful also make it toxic, it needs to be controlled tightly. Additionally, in the reducing environment of the cytosol, Cu is present as insoluble Cu(I). To circumvent both toxicity and solubility issues, a system has evolved where copper is comforted by certain copper binding proteins, so-called Cu-chaperones. They transiently interact with each other to distribute the Cu atoms in a cell. In humans, one of them is Atox1. It binds copper with a binding site containing two thiol residues and transfers it to other binding sites, mostly those of a copper pump, ATP7B (also known as Wilsons disease protein).

My work was aimed at understanding copper-mediated protein-protein interactions on a molecular and mechanistic level. Which amino acids interact with the metal? Which forces drive the transfer from one protein to the other? Using biophysical and biochemical methods such as chromatography and calorimetry on wild type and point-mutated proteins in vitro, we found that the copper is transferred via a dynamic intermediate complex that keeps the system flexible while shielding the copper against other interactions.

Although similar transfer interactions can be observed in other organisms, and many conclusions in the copper field are drawn from bacterial and yeast analogs, we believe that it is important to investigate human proteins, too. Not only is their regulation different, but also only in humans we find the diseases linked to the proteins: Copper level regulation diseases are to be named first, but atypical copper levels have also been linked to tumors and amyloid dispositions. In summary, my observations and conclusions are of basic research character and can be of importance for both general copper and human medicinal research.

After Ctr1-mediated copper ion (Cu) entry into the human cytoplasm, chaperones Atox1 and CCS deliver Cu to P-1B-type ATPases and to superoxide dismutase, respectively, via direct protein-protein interactions. Although the two Cu chaperones are presumed to work along independent pathways, we here assessed cross-reactivity between Atox1 and the first domain of CCS (CCS1) using biochemical and biophysical methods in vitro. By NMR we show that CCS1 is monomeric although it elutes differently from Atox1 in size exclusion chromatography (SEC). This property allows separation of Atox1 and CCS1 by SEC and, combined with the 254/280 nm ratio as an indicator of Cu loading, we demonstrate that Cu can be transferred from one protein to the other. Cu exchange also occurs with full-length CCS and, as expected, the interaction involves the metal binding sites since mutation of Cu-binding cysteine in Atox1 eliminates Cu transfer from CCS1. Cross-reactivity between CCS and Atox1 may aid in regulation of Cu distribution in the cytoplasm.

Although strictly regulated, pH and solute concentrations in cells may exhibit temporal and spatial fluctuations. Here we study the effect of such changes on the stability, structure, and dynamics in vitro and in silico of a two-domain construct (WD56) of the fifth and sixth metal-binding domains of the copper transport protein, ATP7B (Wilson disease protein). We find that the thermal stability of WD56 is increased by 40 °C when increasing the pH from 5.0 to 7.5. In contrast, addition of salt at pH 7.2 decreases WD56 stability by up to 30 °C. In agreement with domain-domain coupling, fractional copper loading increases the stability of both domains. HSQC chemical shift changes demonstrate that, upon lowering the pH from 7.2 to 6, both His in WD6 as well as the second Cys of the copper site in each domain become protonated. MD simulations reveal increased domain-domain fluctuations at pH 6 and in the presence of high salt concentration, as compared to at pH 7 and low salt concentration. Thus, the surface charge distribution at high pH contributes favorably to overall WD56 stability. By introducing more positive charges by lowering the pH, or by diminishing charge-charge interactions by salt, fluctuations among the domains are increased and thereby overall stability is reduced. Copper transfer activity also depends on pH: delivery of copper from chaperone Atox1 to WD56 is more efficient at pH 7.2 than at pH 6 by a factor of 30. It appears that WD56 is an example where the free energy landscapes for folding and function are linked via structural stability.

The diphtheria toxin-like ADP-ribosyltransferases (ARTDs) are an enzyme family that catalyses the transfer of ADP-ribose units onto substrate proteins, using nicotinamide adenine dinucleotide (NAD(+)) as a co-substrate. They have a documented role in chromatin remodelling and DNA repair; and inhibitors of ARTD1 and 2 (PARP1 and 2) are currently in clinical trials for the treatment of cancer. The detailed function of most other ARTDs is still unknown. Using virtual screening we identified small ligands of ARTD7 (PARP15/BAL3) and ARTD8 (PARP14/BAL2). Thermal-shift assays confirmed that 16 compounds, belonging to eight structural classes, bound to ARTD7/ARTD8. Affinity measurements with isothermal titration calorimetry for two isomers of the most promising hit compound confirmed binding in the low micromolar range to ARTD8. Crystal structures showed anchoring of the hits in the nicotinamide pocket. These results form a starting point in the development of chemical tools for the study of the role and function of ARTD7 and ARTD8.

Transient protein-protein and protein-ligand interactions are fundamental components of biological activity. To understand biological activity, not only the structures of the involved proteins are important but also the energetics of the individual steps of a reaction. Here we use in vitro biophysical methods to deduce thermodynamic parameters of copper (Cu) transfer from the human copper chaperone Atox1 to the fourth metal-binding domain of the Wilson disease protein (WD4). Atox1 and WD4 have the same fold (ferredoxin-like fold) and Cu-binding site (two surface exposed cysteine residues) and thus it is not clear what drives metal transfer from one protein to the other. Cu transfer is a two-step reaction involving a metal-dependent ternary complex in which the metal is coordinated by cysteines from both proteins (i.e., Atox1-Cu-WD4). We employ size exclusion chromatography to estimate individual equilibrium constants for the two steps. This information together with calorimetric titration data are used to reveal enthalpic and entropic contributions of each step in the transfer process. Upon combining the equilibrium constants for both steps, a metal exchange factor (from Atox1 to WD4) of 10 is calculated, governed by a negative net enthalpy change of ∼10 kJ/mol. Thus, small variations in interaction energies, not always obvious upon comparing protein structures alone, may fuel vectorial metal transfer.

Metal coordination is required for function of many proteins. For biosynthesis of proteins coordinating a metal, the question arises if the metal binds before, during or after folding of the polypeptide. Moreover, when the metal is bound to the protein, how does its coordination affect biophysical properties such as stability and dynamics? Understanding how metals are utilized by proteins in cells on a molecular level requires accurate descriptions of the thermodynamic and kinetic parameters involved in protein-metal complexes. Copper is one of the essential transition metals found in the active sites of many key proteins. To avoid toxicity of free copper ions, living systems have developed elaborate copper-transport systems that involve dedicated proteins that facilitate efficient and specific delivery of copper to target proteins. This review describes in vitro and in silico biophysical work assessing the role of copper in folding and stability of copper-binding proteins. Examples of proteins discussed are: a blue-copper protein (Pseudomonas aeruginosa azurin), members of copper-transport systems (bacterial CopZ, human Atox1 and ATP7B domains) and multi-copper ferroxidases (yeast Fet3p and human ceruloplasmin). The consequences of interactions between copper proteins and platinum-complexes are also discussed. 

Take a closer look: Unexpectedly, a pair of enantiomeric ligands proved to have similar binding affinities for acetylcholinesterase. Further studies indicated that the enantiomers exhibit different thermodynamic profiles. Analyses of the noncovalent interactions in the protein-ligand complexes revealed that these differences are partly due to nonclassical hydrogen bonds between the ligands and aromatic side chains of the protein.