guest ::
| Home > Publications database > Serial protein crystallography in cellulo |
| Home > Publications database > Serial protein crystallography in cellulo |
| Dissertation / PhD Thesis | PUBDB-2024-05780 |
; ; ;
2024
This record in other databases:
Please use a persistent id in citations: urn:nbn:de:gbv:18-ediss-119789 doi:10.3204/PUBDB-2024-05780
Abstract: Elucidation of high-resolution three-dimensional protein structures via X-ray crystallography requires protein crystals for signal enhancement. Protein crystals have been observed in living cells from all domains of life. This cumulative dissertation refines and expands a systematic approach for crystallization of recombinantly introduced proteins in living insect cells, as a complementary method to conventional crystallization approaches.In particular, detection of crystals in cellulo has been optimized by development of a small angle X-ray scattering/ powder diffraction synergistic method to screen cell cultures. To this end, recombinant baculovirus infected cell cultures were measured at a low-background optimized synchrotron beamline. Their X-ray scattering profile was compared to that of mockvirus-infected and uninfected insect cells. If protein crystals are irradiated, they are in random orientation within the cells. Consequently, weak incomplete Debye-Scherrer rings are recorded, which by summation of multiple detector images, radial averaging, and background subtraction show as peaks in the 1D-plots of the scattering curve. It could be shown that a crystallization efficiency in cells of less than one to six per cent suffices to identify within seconds if there are protein microcrystals within a cell culture. This is largely dependant on the expressed target protein. To this end, proteins were chosen that form clearly visible crystals identifieable by light microscopy. Thus, the efficiency and sensitivity of the established method could be elucidated. Additionally, the dynamic behaviour of crystal growth within the cell ensemble was investigated. This enabled determination of the optimal timepoint after infection for a serial crystallography (SX) experiment.It became clear that data collection within living cells, meaning in cellulo with no prior extraction and purification of crystals, does not increase background noise. Going further, the surrounding cell even protects protein crystals from dehydration and damage during crystal extraction. In this work, SX data collection in cellulo was established at synchrotron radiation sources and using X-ray free-electron laser (XFEL) radiation. With synchrotron radiation, efficient data collection on MicroMeshes (MiteGen) with helical line scans is demonstrated on different constructs of the proteins Neurospora crassa HEX-1 (NcHEX-1) and Tb inosine 5’-monophosphate dehydrogenase (TbIMPDH), solving the structures. In addition, data collection at room temperature (RT) is demonstrated. Using CrystalDirectTM plates with a poly D-lysine coating for enhanced cell adhesion, everything from cell infection, crystal growth up to data collection in the intact cells can be done in situ. This way, any damage on potentially fragile intracellular crystals is prevented. Collecting at RT can also uncover potential alternative conformers. The method enabels automatable data collection on up to 96 sampels without changing the mount.Successful SX in cellulo on XFELs is demonstrated with a fixed-target approach on micro-patterned silicon chips. Compared to liquid jets, risks of nozzle clogging or crystal sedimentation in the sample reservoir are eliminated, reducing sample consumption and increasing hitrates.Interestingly, crystallization inside cells enables binding of genuine co-factors abundant at physiological concentrations. Ligands can either diffuse into or co-crystallize with the crystal. Collecting diffraction data in viable insect cells can lead to identification of such co-factors, provided sufficiently strong binding and data quality. This way it can be shown that the inactive conformation of TbIMPDH incorporates phosphate at the IMP binding site, as well as GDP and ATP in alternate conformations at canonical nucleotide binding sites.Solving the structure after successful data collection can be done by phase retrieval with molecular replacement. Phases can be transfered from a experimentally phased homologue, or more recently from in silico models generated f.i. with AlphaFold2. In order to prevent bias, it is still desirable to introduce heavy atoms into intracellular protein crystals, which later could be used for experimental de novo phasing. Towards this end, TbIMPDH constructs with terminal lanthanide binding tags (LBT) were successfully crystallized in cellulo, unfortunately leading to increased crystal fragility compared to the TbIMPDH in cellulo crystals without LBT. Moreover, unspecific binding of the lanthanide terbium to other moelcules within the cell was confirmed, and, terbium soaking reduced resolution of diffraction data. As an alternate approach, the incorporation of selenium in the form of selenium methionine was tested on different targets during gene expression. It was hypothized and proven by mass spectrometry for NcHEX-1 that a high virus titer (MOI 10) results in an increased selenium incorporation. The potential of this approach for a future phasing via SAD needs to be validated on additional proteins.Results of this cumulative dissertation contribute to the method spectrum of structural biology, opening the approach of in cellulo crystallization to a wider group of researchers. Crystal detection and serial data collection in living cells has been simplified and optimized, rendering the method an alternative crystallization method.
; 
|
The record appears in these collections: |
PDF
PDF (PDFA)