Crystal structure of a bacterial photoactivated adenylate cyclase determined at room temperature by serial femtosecond crystallography
Dr Sofia M Kapetanaki [1]
Molecular Tour
In the search of light-sensing proteins for optogenetic applications, PACs (photoactivated adenylate cyclases) have emerged as very promising candidates in the field, combining the function of a photoreceptor with that of an enzyme capable of a second messenger generation. In order to understand their photoactivation mechanism and enhance the optogenetics toolkit, we need to understand their structural dynamics. Protein conformational changes span a vast range of time scales, from picoseconds to minutes, with ps–ms (picosecond to millisecond) dynamics being particularly relevant to their function. Time-resolved serial crystallography at synchrotrons and XFELs (X-ray free electron lasers) can provide snapshots of the light-induced structural intermediates and hence atomic level information in the time regime of interest. Knowing the dark-adapted state structures of these light-sensing proteins is the starting point for such studies.
In this work, we present the dark-adapted state structures of OaPAC, a recently discovered BLUF (blue-light using flavin adenosine dinucleotide) containing, photoactivated adenylate cyclase from the cyanobacterium Oscillatoria acuminata that uses ATP (adenosine triphosphate) and translates the light signal into the production of cAMP (cyclic adenosine monophosphate). The room temperature structures of the substrate-free OaPAC were obtained for the first time at an XFEL (Image 1) and a synchrotron, and are compared with the cryo-macromolecular structures obtained from data collection from crystals held at around 100 K by us and other researchers. In line with recent observations that SFX (serial femtosecond crystallography) and SSX (serial synchrotron crystallography) measurements can result in data of equivalent quality, our SFX, SSX and cryo-structures of OaPAC do not show any significant differences (Image 2). However, comparison of our OaPAC structures with the recently released OaPAC structures of Chretien et al. and the available OaPAC structures of Ohki et al. reveals some interesting features, presumably arising from the different crystallization conditions used. In particular, we observed a significant reorientation of a helix in the ATP binding domain in our crystallization conditions. This reorientation seems to originate from a loop placed on top of the α5 helix which in our preparation appears to push down the α5 helix resulting in the observed conformation of the helix (Image 3a). As a consequence, the ASA (accessible solvent area) for the ATP pocket is reduced from 511 Å2 in the cryo-structure of Chretien et al. (pdb: 8qfe) (Image 3b) to 315 Å2 in our cryo-structure (pdb: 9f1y) (Image 3c). This is a significant reduction which results in a close conformation that may prevent ATP from binding. These findings highlight the importance of having an open conformation for ATP binding (pdb: 8qfh, ASA: 663 Å2, Image 3d) and the importance of the potential effect of the crystallization conditions.
OaPAC, like all BLUF containing photoreceptors rearranges its hydrogen-bond network (Image 4) around the flavin upon blue light illumination, resulting in a characteristic 10 nm red-shift in the main visible absorption band of the flavin with the signaling state decaying back in seconds to the dark-adapted state. In general, tyrosine 6 plays a key role in the photoactivation mechanism of OaPAC and of BLUF domains, as a PCET (concerted proton-coupled electron transfer) from Y6 which forms the neutral flavin radical takes place. PCET is followed by a recombination of the radical species that results in the hydrogen bond rearrangement around the flavin and transduction of the light signal to the AC domains. To investigate the structural role of this residue, we have replaced Y6 to a tryptophan (Y6W), a mutation which forms a photoinactive enzyme. As anticipated the mutation disturbs the H-bond network around the flavin, resulting in small changes in the H-bond distances of several residues in the flavin environment with the most extensive change being a ~85o degree rotation around the CG – CD bond of the Q48 residue, accompanied by a ~20o rotation of the side chain of M92 (Image 5). Although to a different extent, the observed mutation-induced changes bear a resemblance to the blue light induced changes responsible for the hydrogen bond rearrangement around the flavin in the WT protein. Overall, our study paves the way for time-resolved studies at XFELs (TR-SFX) and synchrotrons (TR-SSX) which will allow a correlation of the available spectroscopic signatures of the substrate-free enzyme with the early-formed structural intermediates, making it possible to obtain a complete picture of the photoactivation mechanism of OaPAC.
References
- ↑ Kapetanaki SM, Coquelle N, von Stetten D, Byrdin M, Rios-Santacruz R, Bean R, Bielecki J, Boudjelida M, Fekete Z, Grime GW, Han H, Hatton C, Kantamneni S, Kharitonov K, Kim C, Kloos M, Koua FHM, de Diego Martinez I, Melo D, Rane L, Round A, Round E, Sarma A, Schubert R, Schulz J, Sikorski M, Vakili M, Valerio J, Vitas J, de Wijn R, Wrona A, Zala N, Pearson A, Dörner K, Schirò G, Garman EF, Lukács A, Weik M. Crystal structure of a bacterial photoactivated adenylate cyclase determined by serial femtosecond and serial synchrotron crystallography. IUCrJ. 2024 Nov 1. PMID:39470573 doi:10.1107/S2052252524010170
