Skip to main content
European Commission logo
English English
CORDIS - EU research results
CORDIS
CORDIS Web 30th anniversary CORDIS Web 30th anniversary

Folding Pathways of DNA G-quadruplexes in Crowding Conditions, and Implications for Mass Spectrometry-based Ligand Screening Assays

Periodic Reporting for period 1 - CROWDASSAY (Folding Pathways of DNA G-quadruplexes in Crowding Conditions, and Implications for Mass Spectrometry-based Ligand Screening Assays)

Reporting period: 2018-06-01 to 2020-05-31

Understanding G-quadruplex (G4) conformations/folding pathway in cellular mimicking conditions is of prime importance. This will help to design ligands for stabilizing a particular topology among different conformations in order to affect G4 mediated biochemical pathways. Until now most of the in vitro biophysically assays were performed in aqueous solutions (with ~100 mM K+) thus did not account for the effect of the cellular crowding conditions (macromolecular crowding and co-solutes). The main aim of the project was to understand whether and how conformation/folding and ligand binding aspects of G quadruplexes differ in dilute aqueous solution and presence of co-solutes.

In this work, on the methodological aspect, we tried to develop native MS of G quadruplex in presence of co-solutes (compatible with electrospray ionization). Then we demonstrated the conformational transition of different topologies of quadruplexes in the presence of two different co-solutes with a focus on telomeric G-quadruplexes (G4). Next, we screened different telomeric G4 sequences against two well-known G4 binders (PhenDC3 & 360A) and found one lead complex [23TAG (PDB: 2JSM) & PhenDC3 complex] for further in-depth structural characterization by solution NMR (Secondment in Slovenia). It is noteworthy that in the literature there are no high-resolution structures of ligand-bound 2-quartet G4. Therefore, in this study, we showed for the first time the atomistic details of ligand bound-2 quartet antiparallel human telomeric G-quadruplex in K+ solution.

Further, we tried to push the limits of in vitro screening of G4 specific ligands by adding the co-solutes in the solution to make them more relevant to the cellular condition. The differential binding modes/stoichiometry of ligands in presence of co-solutes pinpoints that current screening strategies in dilute aqueous solutions are inadequate, which needs to be revised by taking into account the co-solutes.

Another advantage of our approach was to integrate orthogonal solution spectroscopic techniques (CD, 1HNMR) with native MS (coupled with ion mobility spectrometry) to give complementary information on the different co-existing states and their structures. We incorporate them in a unified manner to provide a robust model of G-quadruplex folding.

In conclusion, this work contributed to reveal some fundamental principles of nucleic acid folding.
To develop native mass spectrometry in crowding conditions (WP1) & understand G-quadruplex folding in crowding conditions (WP2):
We tested 19 G-quadruplex (G4) sequences of diverse lengths (19-26nt), diverse topology and origin in native MS conditions (for details see Table 1 of Technical Part report). We found two co-solutes (Ethanol and Acetonitrile) that are compatible with electrospray ionization with a concentration range of 10-60% v/v. Then we used the co-solutes at different concentrations (10%-60%) against the quadruplex sequences to understand the effect on structure and folding (D1.1+D1.2)
For Parallel 3-quartet G4 sequences we did not observe any notable change in K+ stoichiometry up to a 50% v/v of acetonitrile & ethanol. Interestingly for Antiparallel 2-quartets we observed a remarkable change in K+ stoichiometry from 2-quartet to 3-quartet in ESI MS.
Complementary solution spectroscopy i.e. CD & NMR showed the conformational reorganization to parallel conformation from antiparallel/ hybrid conformation in the presence of co-solutes (D1.3).
Next, to understand the kinetic aspects of co-solute induced conformational rearrangement we focused on polymorphic telomeric G-quadruplexes (TTAGGG)n in K+ solution (D2.1). We performed time-resolved mass spectrometry in the presence of 50% v/v co-solutes (Ethanol & Acetonitrile) to distinguish folding intermediates based on the number of cations and their rates of appearance. It is complemented by solution spectroscopy i.e. CD to characterize the folding ensembles based on their secondary structure. Indeed, a multi spectroscopy/spectrometry-based approach has been performed to understanding a unified folding pathway in co-solute induced crowding conditions.
G4-ligand assays in absence/presence of co-solutes (WP3):
For MS based ligand-binding assays we selected two well-known G4 binders (PhenDC3 & 360A). Then we tested these two ligands against the six different variants of HT G4 sequences (D3.1+3.2). We found both the ligands were capable of confiscating one cation from the 3-quartet species accompanied by conformational rearrangement to antiparallel from the hybrid conformation. At higher ligand-DNA molar ratio we observe more than one ligand is binding to one G4 sequence. Complementary solution spectroscopy i.e. CD in MS buffer and high salt containing buffer showed the conformational rearrangement to antiparallel conformation from hybrid conformation. We also did detailed solution NMR titrations with these two ligands to find any suitable candidates for structural characterization by NMR. Interestingly the appearance of sharp NMR signals (in the imino region) of 23TAG (PDB: 2JSM, 3 quartets, Hybrid) in presence of 1:1 molar equiv. of PhenDC3 intrigued us to investigate the structural details of the complex by NMR. Therefore, I carried out a secondment in the Slovenian NMR center with Prof. Janez Plavec’s group. From the combined analysis of the several 2D homo-nuclear and heteronuclear experiments, we projected the Schematic model of the complex which is antiparallel 2-tetrad G quadruplex with chair conformation in its ligand-bound state (See Figure 1 of Technical Part report).
Next, MS-based screening of the same set of sequences/ligands in the presence of co-solutes (50% v/v) showed a differential binding behavior (D3.3). For the 2-quartet antiparallel G4s we observed formation of 3 quartet complexes along with 2 quartet complexes for both ligand bound states. For hybrid conformations we also observed the same trend along with presence of ligand free 3 quartet species which showed reduced affinity in presence of co-solutes. The solution CD in MS condition supports the MS results in terms of mixed populations of different species. Therefore, we revealed that current screening strategies in dilute aqueous solutions need to be reviewed by considering the effect of co-solutes.
We have shown state-of-the-art native MS conditions for studying G-quadruplexes in K+ solutions in presence of co-solutes. In a broader application these conditions can be used for studying other types of nucleic acid (Duplex DNA & RNA, Riboswitches, etc.) and intrinsically disordered protein (IDP) folding in MS-compatible volatile buffer. Therefore, the native MS conditions can be a starting point for the study of protein folding and aggregation in presence of co-solutes.
On the other hand, during secondment it created strong collaboration between two European institutes (IECB in France and the NIC in Slovenia). In the future, this will build a long-term collaboration between those to solve complex scientific problems and thus an exchange of researchers.
Effect of co-solute(acetonitrile) on the binding of PhenDC3 to Human Telomeric hybrid G-quadruplex