Overview

1) Channels labelling

Many enzymes have their active sites deeply buried within their core connected with the solvent by channel networks. These channels are used to transport the substrates from the solvent to the catalytic centers, and export the products back to the solvent. Tunnels have evolved to filter-out inhibitors, or select the good substrates enantiomer (sterochemistry). Hence, these channel networks play are crucial for the enzymes' functions. This traffic regulation process is present in all the 6 classes of enzymes (figure 1, depicted from Prokop, Z. et al. Engineering of Protein Tunnels. Engineering Handbook. Wiley; 2013. p. 421-464.). Importantly, the structure and architecture of these tunnels need to be deciphered,  to understand an important part of the enzymes mechanisms. The understanding of tunnels functions can also help designing more effective enzymes or better drugs.

Figure 1. Internal channel networks in enzymes. Channels are crucial for their for ezymes functions.

 

To decipher at atomic resolution the architecture of protein channels in crystallo, we have design an effective cryogenic krypton pressure cell (figure 2). Users who have crystallized enzymes with internal channels, and who are interested in studying the potential struture of a or several channel networks of their enzymes are welcome to use our system.

 

Figure 2. On the right, a picture of the krypton pressure cell in the HPMX lab. On the left, a schematic represenation of the main components of the system.

 

2) Structure phasing.

Krypton labelling can also be used for phasing to solve structures de novo. The K absorption edge of Kr can be found at an energy of 14.3 keV (figure 3), well within the energy range of tuneable macromolecular crystallography beamlines (e.g. ID23-1, ID30B). It allows for both multiple-wavelength anomalous diffraction (MAD) phasing experiments around the K edge of Kr for an unequivocal identification of Kr binding sites. Krypton is neutral, i.e. it does not chemically interact with protein, is bound via London forces, and sits preferably in non-polar or hydrophobic cavities in proteins (but not exclusively). The fact that the anomalous scattering factor (f") has 3.8 electrons at the absorption edge, provides krypton with a sufficiently high phasing power. Users who crystallized new proteins whith no available homologous structures, and in need of a good derivative, are welcome to use our system.

 

 

Figure 3.  Krypton Ka-absorption edge mesured by XRF on MX-beamline.

 

Experimental setup

We have design a standalone pressure cell for soaking biological crystal in pressurized krypton prior to cryo-cool them (figure 4). The concept is based on the phase diagram of krypton, which is in a gas phase at ambient temperature and in a liquid phase at cryogenic temperatures (from 120 K). After the crystal is harvested and cryoproteted, it is placed in the top part of a vertical pressure-tube, pressurized in krypton gas at ambient temperature (krypton atoms bind to the protein and populate the channels). After some time, the sample is released from the top part to drop into liquid krypton where it is frozen. After returning to ambient pressure, the fozen and stable sample is handled further at cryogenic temperature. Advantageously, the samples are pressurize at 150 bar in pure krypton. The protein-krypton complex is cryo-cooled under pressure at the equilibrium where the binding process is at the optimum.

Figure 4. Block diagram of the cryogenic krypton pressure cell.

 

 

Examples

1)  Structure of hydrogenase Kr-derivative and gas pathways

NiFe hydrogenases enzymes catalyse the splitting of molecular hydrogen into protons, and are interesting for developments in renewable energy technologies. Generally, NiFe hydrogenases are irreversibly inactivated by O2, but few of them appear to be tolerant to atmospheric levels of oxygen, and are then more interesting for applications. The hydrogenase of Ralstonia eutropha (ReMBH) falls into this category. Catalysis takes place at the NiFe active site, which is deeply buried within the protein and connected to the surface via hydrophobic channels. These are responsible for the transport of both substrate and inhibitory gases. However, the pathways for the gas molecules (H2 and O2) were largely unknown. To visualise the tunnels network, krypton derivatised crystals were produced using our cryogenic krypton pressure cell. The crystals were pressurised for 10 minutes under 80 bar of krypton gas, and then flash-cooled under pressure. X-ray diffraction data were collected on beamline ID23-1. 19 krypton sites were identified inside the protein, which map a hydrophobic gas tunnel network that connects the protein surface with the NiFe catalytic centre (figure 5). Calculations revealed considerable differences between O2-tolerant and O2-sensitive hydrogenases both in tunnel size and tunnel quantity. O2-sensitive NiFe hydrogenases have on average twice as many hydrophobic gas channels as the O2-tolerant ones. In the present O2-tolerant hydrogenases, the existence of only two gas channels may allow for a much tighter control of the flow of gas molecules to the sensitive catalytic centre, therefore explaining their O2 tolerance. This result was published in J. Kalms et al. (2016). Angew. Chem. Int. Ed. 55, 5586-5590.

Figure 5.  Crystallographic structure of the Kr-derivative of hydrogenase from Ralstonia eutropha (ReMBH).

 
 
2)  Mapping hydrophobic tunnels and cavities in neuroglobin using krypton labelling

Neuroglobin (Ngb) is a neuronal hexacoordinated protein of the globin family, likely involved in the protection of neurons in vertebrates. The internal heme cavities of Ngb play a functional role in binding small gas molecules. Noble gas labeling studies revealed that Xe binds to four distinct sites at 20 bar (diffraction data collection at 100 K) and to seven sites when pressurised at 40 bar (diffraction data collection at room temperature). Two Xe binding sites are located within the large internal heme cavity (“figure 6, storage” and “relay” site) were proposed to store transiently gaseous molecules (e.g. CO, NO, O2 …) in transit to the heme.

Figure 6.  Crystallographic structures of Kr-derivatives of neuroglobin.

 
To map a diffusion pathway within the internal cavity of Ngb, we determined a WT-Ngb structures under 100 bar and 150 bar of krypton pressure, respectively. The studies suggest a new diffusion pathway for oxygen. The structure at 150 bar shows a continuous chain of krypton atoms that trace a channel network with 3 entries for O2 and 3 branches that converge to the heme-group (figure 7). The main route for oxygen is as follows: pocket (entry) -> storage cavity-> relay cavity -> Fe. This result has been published in N. Colloc’h et al. (2017). Biophysical journal 113.
 
 

Figure 7.  Hydraphobic channels in neurogobin.