Methods and tools for the determination of native structural information of biomolecules using hypenating mass spectroscopy

Abstract

Method for the determination of the conformation of at least one biopolymer by using hyphenating mass spectroscopy with chromatographic or electrophoretic separation, wherein a) downstream of the chromatographic or electrophoretic separation electrospray ionization (ESI) is used for generating native or essentially native conformation ions of said of at least one biopolymer in the gas phase; b) these ions, or precursors based on these ions, are selected, preferably based on at least one of mobility, charge state, or a combination thereof, c) these selected ions or precursors are fragmented using a non-ergodic or essentially non-ergodic dissociation to form fragments; d1) the resulting fragments are on the one hand analyzed for their distribution as a function of mass to charge in a survey spectrum, preferably using a selection based on at least one of mobility, charge state, or a combination thereof; d2) the resulting fragments are on the other hand collision activated and then analyzed for their distribution as a function of mobility to charge in at least one spectrum as a function of said collision activation, preferably analyzed using a selection based on at least one of mobility, charge state, or a combination thereof, e) using differential information from d1) and d2) about the fragments to determine which of the fragments resulting from step c) are in a native or essentially native conformation, and using information from the fragments having native conformation for the determination of the conformation of said at least one biopolymer.

Description

[0001] TITLE

[0002] METHODS AND TOOLS FOR THE DETERMINATION OF NATIVE STRUCTURAL INFORMATION OF BIOMOLECULES USING HYPENATING MASS SPECTROSCOPY

[0003] TECHNICAL FIELD

[0004] The present invention relates to methods for the determination of native structure of biomolecules, in particular proteins, with a focus on determining secondary, tertiary and quaternary structural information, as well as to devices for such methods.

[0005] PRIOR ART

[0006] Proteoforms, the molecular forms of an expressed protein, establish the molecular basis of cellular function. Because proteoforms can differ in their structural dynamics, changes in proteoforms can alter the structure or formation propensity of protein assemblies, alter protein-protein interaction networks, and lead to disease states. Typical mammalian cells are projected to comprise one million proteoforms, of which only a minuscule number has currently been identified, and which can vary over time. This complexity of the proteome at all levels of protein structure and dynamics renders the systematic identification of how variations in proteoforms alter cellular protein-protein interaction networks and cause disease a formidable, unsolved challenge.

[0007] Mounting evidence suggests that electrospray ionization-mass spectrometry largely retains native structures of proteins and protein complexes in the absence of bulk solvent. Specifically, hyphenating electrospray ionization-mass spectrometry with ion spectroscopy, ion mobility spectrometry, or cryogenic electron microscopy revealed that solvent-free measurements can be carried out on a timescale that is shorter than the time required for the protein to denature due to the lack of bulk solvent. Because of their high sensitivity and throughput, the ability to retain native-like protein structures potentially allows mass spectrometry-based methods to screen proteomes for structural changes to identify on a proteome-wide scale how differential protein modifications alter protein structures, misregulate protein-protein interaction networks, and cause disease. To date, however, proposing complete protein structures directly from mass spectrometry-based measurements has not yet been accomplished because the acquired data do not contain sufficient structural information. Elucidating the complete structure of a protein has the additional requirement that structural information be directly available from the experimental data. To date, however, mass spectrometry-based methods provide insufficient structural information to propose a complete structure for a protein. Hyphenating mass spectrometry with ion mobility spectrometry has enabled measurements of orientationally-averaged collision cross-sections of protein systems. Although collision cross-sections are valuable because they directly reveal the overall compactness of the protein system, an individual collision cross-section does not distinguish between different protein conformations that are similarly compact.

[0008] US 7,838,826 B1 (M. A. Park, 2008) and the corresponding patent family members presents a small ion mobility analyzer / spectrometer which has become known under the acronym “TIMS” analyzer / spectrometer (TIMS = trapped ion mobility spectrometry). The terms ion mobility analyzer and ion mobility spectrometer are used interchangeably here. A TIMS analyzer comprises a gas flow that drives ions against a counter-acting electric field barrier such that the ions are at first trapped along the axis of the TIMS analyzer. The ions are confined in the radial direction by an electric RF field. After transferring ions from an ion source to the electric field barrier, the height of the electric field barrier or the gas velocity is adjusted such that ion species are released from the electric field barrier in the sequence of their mobility.

[0009] Commonly, the length of the ion mobility separation unit of a TIMS analyzer amounts to about five centimeters only. In a small tube with an inner diameter of about eight millimeters, a radial RF quadrupole field is generated to hold ions near to the axis. A gas flow inside a tube drives ions entrained in the gas flow against a ramped counter-acting electric DC field barrier where the ions are trapped and separated according to their mobilities at locations on the field ramp at which the friction force of the moving gas equals the counter-acting force of the electric DC field on the ramp. After loading the TIMS with ions, the height of the electric DC field barrier is decreased; this scan releases the ion species in the sequence of their mobility. Unlike many other trials to build small ion mobility spectrometers, the small device by M. A. Park has already achieved, with reduced scan speeds, ion mobility resolutions up to RmOb = 400, which is extraordinarily high.

[0010] Regarding the theoretical basis of TIMS, see the research article “Fundamentals of Trapped Ion Mobility Spectrometry”, K. Michelmann, J. A. Silveira, M. E. Ridgeway and M. A. Park, J. Am. Soc. Mass Spectrom., (2015) 26: 14-24.

[0011] US 8,766,176 B2 proposes improvements of the scan modes for TIMS analyzers by application of non-linear scans to achieve a linear mobility scale, a constant resolution along the mobility scale, or a temporal zoom. Furthermore, US 9,984,864 B2 describes a spatial zoom.

[0012] A TIMS analyzer with parallel ion accumulation is described in US 9,683,964; it improves the utilization of the ions from the ion source to nearly 100%. Importantly, many ion sources, such as electrospray ion sources produce a continuous ion beam. TIMS with parallel accumulation can also be operated to collect and analyze ions continuously - that is, TIMS can operate at 100% duty cycle. Barring pseudopotential or space charge effects, substantially all ions of the ion source are collected and analyzed without loss. TIMS with parallel ion accumulation further provides the possibility to prolong the ion accumulation and scan duration, thereby increasing the ion mobility resolution so as to separate and detect more ion species.

[0013] The ions are collected in an accumulator unit, preferably almost identical to the scanning unit, at a ramp of an electric DC field barrier such that they get spatially separated by their ion mobility along the ramp. Therefore, the accumulated ions are less influenced by space charge than in other types of accumulator units. Of greatest importance, however, is the unique feature of a TIMS analyzer that a longer accumulation period permits to increase the mobility resolution by choosing correspondingly longer mobility scan durations, e.g. 100 milliseconds scan duration with an ion mobility resolution of RmOb = 75 instead of 20 milliseconds scan duration with RmOb = 30. As a consequence of the higher number of ions collected and the better ion mobility resolution, more ion species can be detected and measured. Once an ion mobility scan is completed (optionally after twenty to some hundred milliseconds), the accumulated ions are transferred (in about a millisecond) from the accumulation unit to the scanning unit, and the next ion mobility scan can be started. In total, a skilled practitioner will appreciate that it will be possible to achieve a measurement rate of 300 to 450 ion species per second.

[0014] Some improvements for higher amounts of stored ions in selected regions of ion mobility, particularly for ions of low ion mobility, are given in US 9,304,106 B1. The higher loading capacity is based on non-linear electric DC field ramps, with flatter field ramps for ion species of interest, in order to diminish the effect of space charge for these ion species. But for precise ion mobility analyses of low abundant ion species in complex mixtures the influence of the space charge is still high.

[0015] Fig. 1 depicts a prior art, mobility-based filtering method according to Park et al. in US 2012 / 0273673. According to this document two mobility devices are provided. Ions to be filtered, entrained in a gas, are introduced at an entrance of the first mobility device, “...gas drives entrained ions against a first field barrier 50, keeping back all ions with mobilities |m[> / =]pL1. This case is schematically illustrated in Fig. 1 showing the disposition of ions 52 to 59 with different mobilities, indicated by different sizes of the dots representing the ions with their cross sections. The gas drives the ions 55 to 59, that have passed the first barrier 50 against a second field barrier 51 , keeping back the ions 55 and 56 with mobilities p> / =p.2and thus keeping back and collecting all ions of a mobility in the range Ap between and ,2with i > .2. The ions 57 to 59 with ,< ,2pass the second barrier and disappear. To collect the ions 55 and 56, there has to be an ion storage device between the filters, e.g. by the provision of radial forces to keep the ions within the collection volume between the two barriers, such like a multipole field with its centripetally acting pseudopotential. The simplest way to generate the storage volume may be an enclosure of both filters in an RF multipole device, for instance, in an RF quadrupole system.”

[0016] US-A-2018340910 proposes a mass spectrometer comprising two ion mobility analyzers in tandem arrangement, of which at least one is a trapped ion mobility spectrometer (TIMS), and an ion gate which is located between the two ion mobility analyzers, and use thereof wherein ions are selectively transferred between the two ion mobility analyzers by adjusting the transmission of the ion gate while ions are separated in time according to ion mobility in the first ion mobility analyzer.

[0017] In sum, the analysis of samples, especially protein samples for primary structure, identification, and quantitation, has been greatly advanced over previous methods, by the hyphenation of IMS with MS. The determination of secondary, tertiary, and quaternary structures, however, still relies on much slower and more expensive chemical-based methods. The measurement by ion mobility mass spectroscopy (IMMS) of the cross section of a protein, as mentioned above, provides a measure of its compactness, however, this is insufficient to determine secondary and tertiary structure.

[0018] Liu et al in J Am Soc Mass Spectrom 2023 Oct 4; 34(10):2232-2246 (doi: 10.1021 / jasms.3c00187) report that "Top-down" proteomics analyzes intact proteins and identifies proteoforms by their intact mass as well as the observed fragmentation pattern in tandem mass spectrometry (MS / MS) experiments. Hybrid ion mobility spectrometry-mass spectrometry (IM / MS) methods have gained traction for top-down experiments, either by allowing top-down analysis of individual isomers or alternatively by improving signal / noise and dynamic range for fragment ion assignment. They recently described the construction of a tandem-trapped ion mobility spectrometer / mass spectrometer (tandem-TI MS / MS) coupled with an ultraviolet (UV) laser and demonstrated a proof-of-principle for top-down analysis by UV photodissociation (UVPD) at 2-3 mbar. The work builds on this with an exploration of a top-down method that couples tandem-TIMS / MS with UVPD and parallelaccumulation serial fragmentation (PASEF) MS / MS analysis. They first survey types and structures of UVPD-specific fragment ions generated in the 2-3 mbar pressure regime of our instrument. They observe UVPD-induced fragment ions with multiple conformations that differ from those produced in the absence of UV irradiation. Subsequently, they discuss how MS / MS spectra of top-down fragment ions lend themselves ideally for probabilitybased scoring methods developed in the bottom-up proteomics field and how the ability to record automated PASEF-MS / MS spectra resolves ambiguities in the assignment of top- down fragment ions. Finally, they describe the coupling of tandem-TIMS / MS workflows with LIVPD and PASEF-MS / MS analysis for native top-down protein analysis.

[0019] Fouque et al in Anal. Chem. 2022, 94, 44, 15377-15385

[0020] (htps: / / doi.org / 10.1021 / acs.analchem.2c03147) report that post-translational modifications (PTMs) on intact histones play a major role in regulating chromatin dynamics and influence biological processes such as DNA transcription, replication, and repair. The nature and position of each histone PTM is crucial to decipher how this information is translated into biological response. In the work, the potential of a novel tandem top-“double-down” approach— ultraviolet photodissociation followed by mobility and mass-selected electron capture dissociation and mass spectrometry (UVPD-TIMS-q-ECD-ToF MS / MS)— is illustrated for the characterization of HeLa derived intact histone H4 proteoforms. The comparison between q-ECD-ToF MS / MS spectra and traditional Fourier-transform-ion cyclotron resonance-ECD MS / MS spectra of a H4 standard showed a similar sequence coverage (~75%) with significant faster data acquisition in the ToF MS / MS platform (~3 vs ~15 min). Multiple mass shifts (e.g., 14 and 42 Da) were observed for the HeLa derived H4 proteoforms for which the top-down LIVPD and ECD fragmentation analysis were consistent in detecting the presence of acetylated PTMs at the N-terminus and Lys5, Lys8, Lys12, and Lys16 residues, as well as methylated, dimethylated, and trimethylated PTMs at the Lys20 residue with a high sequence coverage (~90%). The presented top-down results are in good agreement with bottom-up TIMS ToF MS / MS experiments and allowed for additional description of PTMs at the N-terminus. The integration of a 213 nm UV laser in the present platform allowed for LIVPD events prior to the ion mobility-mass precursor separation for collision-induced dissociation (CID) / ECD-ToF MS. Selected C3o5+LIVPD fragments, from different H4 proteoforms (e.g., Ac + Me2, 2Ac + Me2 and 3Ac + Me2), exhibited multiple IMS bands for which similar CID / ECD fragmentation patterns per IMS band pointed toward the presence of conformers, adopting the same PTM distribution, with a clear assignment of the PTM localization for each of the C3o5+LIVPD fragment H4 proteoforms. These results were consistent with the biological “zip” model, where acetylation proceeds in the Lys16 to Lys5 direction. This novel platform further enhances the structural toolbox with alternative fragmentation mechanisms (LIVPD, CID, and ECD) in tandem with fast, high-resolution mobility separations and shows great promise for global proteoform analysis.

[0021] SUMMARY OF THE INVENTION

[0022] It is therefore a purpose of the present invention to provide a more rapid, less expensive, instrument-based, hardware, and method for obtaining information regarding the secondary, tertiary, and / or quaternary structure of proteins.

[0023] The top-down analysis of protein samples by mass spectrometry-based methods begins with intact proteins and is generally aimed at determining protein primary sequence, post- translational modifications, quantitation, and the like. Top-down analysis of intact proteins and protein complexes by tandem mass spectrometry cleaves protein backbone bonds and produces protein fragments. However, little is known, in the prior art, of the conformation of the fragment ions or the relation of such conformation to that of the secondary or tertiary structure of the precursor protein. If the fragments are produced non-ergodically, i.e. on a time-scale that is shorter than the time-scale of structural rearrangements, via the absorption of UV photons or the capture of electrons, the conformation of the fragment may correlate with that of the intact protein. Specifically, the cleavage propensities of amide bonds are sensitive to the intact protein structure and the structural flexibility of the amino acid residues. If these protein fragments retain their native structures and non-covalent contacts, they can be used to characterize the protein’s internal structure.

[0024] If proteins can be dissociated into fragments while essentially preserving their native conformation, then IMMS methods might be used to characterize these fragments, and by extension, the internal structure of a protein.

[0025] However, the conformations and non-covalent contacts of protein fragments produced by top-down tandem-mass spectrometry analysis are essentially uncharacterized, primarily because of technical limitations in creating, stabilizing, and structurally characterizing them. Consequently, although it has been widely recognized that protein fragments can be informative of the intact protein structure, it remains untested if their structures can be helpful in elucidating the intact protein structure. Although it is established that electrospray ionization mass spectrometry methods essentially retain native protein structures, it is not known from prior art how to utilize the available information to propose atomic-level structural models for protein systems.

[0026] The supposition that intact, native-like protein ions can be non-ergodically fragmented, that the resultant fragment ions can be stabilized and kinetically trapped in a conformation correlated to the precursor, and that the fragment ions can be further analyzed by IMMS methods to produce information about the secondary, tertiary, and quaternary structure of the precursor protein, forms the basis for this invention.

[0027] According to a first aspect of the present invention, it relates to a method for the determination of the conformation of a biopolymer by using hyphenating ion mobility mass spectroscopy, optionally downstream from a chromatographic or electrophoretic separation. According to a first preferred method, a) a hyphenated ion mobility mass spectrometer is provided, including at least one ion mobility analyzer and at least one mass analyzer wherein at least one of the ion mobility analyzers is preferably a trapped ion mobility analyzer (TIMS) and wherein at least one of the mass analyzers is preferably a time-of-flight (TOF) mass analyzer; b) gas phase analyte ions are formed from sample material by electrospray ionization; c) precursor ions of a known mobility are selected from the gas phase analyte ions, preferably via ion mobility analysis, most preferably via TIMS analysis, wherein precursors ions are allowed to pass an ion optical gate as they elute from the mobility analyzer whereas substantially all other ions are not; d) these selected ions or precursors are fragmented using a non-ergodic or essentially non- ergodic method, preferably ultraviolet photodissociation (LIVPD), to form fragment ions; and e) the resultant fragment ions are analyzed by ion mobility - mass spectrometry, preferably TIMS-TOF.

[0028] The terminology "conformation" of at least one biopolymer such as a protein, nucleotide or derivatives thereof is to be understood broadly as commonly accepted in the field, meaning the arrangement in space of the biopolymer's (e.g. protein's) constituent atoms determining the overall shape of the molecule. In other words, the expression "conformation" includes any kind of information going beyond the information of the primary structure, i.e. the linear sequence of building blocks (e.g. amino acids and potential chemical modifications thereof in a peptide or protein). The expression "conformation" therefore includes the secondary structure (three dimensional arrangement of local segments e.g. of proteins, the two most common secondary structural elements are alpha helices and beta sheets, but this also includes beta turns and omega loops), the super-secondary structure (motives, compact three-dimensional protein structure of several adjacent elements of a secondary structure that is smaller than a protein domain or a subunit), the tertiary structure (domains, three dimensional shape e.g. of the protein) and the quaternary structure (structure of proteins which are themselves composed of two or more smaller protein chains, also referred to as subunits).

[0029] The terminology "hyphenating mass-spectrometry" is to be understood as techniques combining (liquid or gas) chromatographic or electrophoretic and / or ion mobility spectrometry and mass spectrometry methods to exploit the advantages of each. Chromatography produces pure or nearly pure fractions of chemical components in a mixture. Mass spectrometry produces selective mass to charge information for identification using standards or library spectra.

[0030] In alternate embodiments, any combination of one or more known ion mobility analyzer and one or more known mass analyzer may be used including drift cell, differential, travelling wave, or TIMS mobility analyzers, and quadrupole, ion cyclotron resonance, orbitrap, time- of-flight, or Paul trap mass analyzers.

[0031] In alternate embodiments, any known method of gas phase ion production including matrix assisted laser desorption ionization, atmospheric pressure chemical ionization, photo ionization, desorption electrospray ionization, or any other known method of ionization. The biopolymer may be any type of biopolymer, including assemblies, including transiently populated assemblies, between proteins and any other biomolecule like RNA, DNA, lipids, etc., but also include synthetic compounds like drugs or drug candidates or fragments of these compounds, individual proteins, peptides, oligonucleotides, or carbohydrates.

[0032] In alternate embodiments, the compounds in the sample material are separated from each other before the ionization step, preferably using separation techniques known to maintain native conformations in biochemistry and those used in top-down analysis. This includes size-exclusion chromatography (SEC), gel-based separations (like SDS-PAGE), capillary electrophoresis (CE), liquid chromatography (LC), and related approaches. In further alternate embodiments, the precursors are separated from other analyte ions and selected based on a combination of the precursors’ mass, mobility, and charge state.

[0033] In a further alternate embodiment, the precursor ions may be trapped and accumulated in an ion trap, preferably an rf ion trap. Preferably the ion trap is comprised of a set of conducting electrodes defining a volume in which ions the ions may be trapped. A radio frequency potential is applied to the electrodes which in turns produces a pseudopotential which confines the ions in the volume. Preferably the ion trap is filled with a gas of sufficient pressure to dampen the motion of the ions. The gas can be any combination of nitrogen, argon, helium, carbon dioxide, or any other inert gas. The static pressure of the gas is preferably about 3 mbar, but alternatively may be any pressure between about 1 E-2 and about 50 mbar.

[0034] In alternate embodiment methods, ion fragmentation is carried out in the ion trap. In further alternate embodiments, fragmentation of the precursor ions is carried out while the precursors are being accumulated in the ion trap or alternatively once a desired group of precursor ions has been accumulated. In a further alternate embodiment, the non-ergodic fragmentation method may be any known method including any photodissociation method - especially high energy photodissociation using photons of UV or shorter wavelengths - or any electron-based fragmentation method such as electron transfer dissociation (ETD), electron capture dissociation (ECD), or electron activated dissociation (EAD). In a further alternate embodiment, the fragment ions are rapidly quenched following the fragmentation step. “Rapidly” in present context, means “on a much shorter time scale than needed for the ion to change conformation”. This time scale is poorly explored in the scientific literature, some authors suggest an upper limit of milliseconds. “Quenched” means “to remove excess energy remaining from the fragmentation process”. It is known by those skilled in the art that collisions with a rest gas in an rf ion trap will cool the ions. Thus, in a preferred embodiment, the fragment ions are cooled by collisions with inert gas in the rf ion trap in which the fragmentation step takes place. In a further alternate embodiment, at least some of the fragment ions are kinetically trapped in a conformation; and wherein “kinetically trapped” means the conformation of the ions is a local minimum energy gas phase conformation but not a global minimum.

[0035] In a further alternate embodiment, the fragment ions need not be mass analyzed, but are only mobility analyzed in step e). In yet further alternate embodiments, the fragment ions are activated, preferably via energetic collisions with an inert gas (Clll), sufficiently to cause at least some of the fragment ions to unfold or change conformation. Alternatively, activation towards unfolding can be via any know ergodic method of activation including ion-molecule collisions, infrared activation, or excitation in a radio frequency field.

[0036] In another alternate embodiment, the fragment ions are analyzed according to any known prior art IMMS method including parallel accumulation serial fragmentation (PASEF), dia PASEF, midia PASEF, and / or dmf PASEF.

[0037] According to second preferred method, a) a hyphenated ion mobility mass spectrometer is provided, including at least one ion mobility analyzer and at least one mass analyzer wherein at least one of the ion mobility analyzers is preferably a TIMS analyzer and wherein at least one of the mass analyzers is preferably a TOF mass analyzer; b) gas phase analyte ions are formed from sample material by electrospray ionization; c) precursor ions of a known mobility are selected from the gas phase analyte ions, preferably via ion mobility analysis, most preferably via TIMS analysis, wherein precursors ions are allowed to pass an ion optical gate as they elute from the mobility analyzer whereas substantially all other ions are not; d) fragment ions are formed, non-ergodicly, from the selected precursor ions preferably via UVPD; e) the resultant fragment ions are analyzed by ion mobility - mass spectrometry, preferably TIMS-TOF; f) steps b) - e) are performed wherein the precursor ions have a first conformation of a target species; g) steps b) - e) are performed wherein the precursor ions have a second conformation of the target species; and h) the results of steps f) and g) are differentially compared to one another.

[0038] One may thus select a first precursor having a mobility indicative of an elongated conformation, for fragments, measure mobility of the fragments; and in a second measurement, select a second precursor of the same protein have a mobility indicative of a compact, presumably more native conformation, form fragments, measure mobility of the fragments, and compare the fragments of the second precursor to that of the first to determine which of the fragments are representative of the native protein conformation.

[0039] In an alternate embodiment, the first conformation is a substantially native conformation of the target species and the second conformation is a substantially elongated, non-native conformation of the target species.

[0040] The above assumes that a compact species is native and that an extended species is nonnative. That should be the case for globular protein systems, but it does not need to be generally true for all possible proteins or all possible protein modifications / protein complexes, etc. The native structure can also be more extended than the non-native structure and regardless of the compactness of the native structure, the proposed method is able to detect the cross-section of the native structure.

[0041] Here “native conformation” or “native structure” is meant to include the entire ensemble of the biopolymer’s conformations which occur under native-like conditions, including low- abundant or transiently populated conformations and so-called hidden states. The target species may be ions formed from any biopolymer including a protein, peptide, oligonucleotide, or carbohydrate. The differential results of step h) may be used to determine which of the fragment ions are more representative of the native target species conformation. The differential results of step h) may further be used to determine the native conformation of the target species or limit its possible native conformations.

[0042] In alternate embodiments, any combination of one or more known ion mobility analyzer and one or more known mass analyzer may be used including drift cell, differential, travelling wave, or TIMS mobility analyzers, and quadrupole, ion cyclotron resonance, orbitrap, time- of-flight, or Paul trap mass analyzers.

[0043] In alternate embodiments, any known method of gas phase ion production including matrix assisted laser desorption ionization, atmospheric pressure chemical ionization, photo ionization, desorption electrospray ionization, or any other known method of ionization.

[0044] In alternate embodiments, the compounds in the sample material are separated from each other before the ionization step preferably using separation techniques known to maintain native conformations in biochemistry and those used in top-down analysis. This includes SEC, gel-based separations (like SDS-PAGE), CE, LC, and related approaches. In further alternate embodiments, the precursors are separated from other analyte ions and selected based on a combination of the precursors’ mass, mobility, and charge state.

[0045] In a further alternate embodiment, the precursor ions may be trapped and accumulated in an ion trap, preferably an rf ion trap. Preferably the ion trap is comprised of a set of conducting electrodes defining a volume in which ions the ions may be trapped. A radio frequency potential is applied to the electrodes which in turns produces a pseudopotential which confines the ions in the volume. Preferably the ion trap is filled with a gas of sufficient pressure to dampen the motion of the ions. The gas can be any combination of nitrogen, argon, helium, carbon dioxide, or any other inert gas. The static pressure of the gas is preferably about 3 mbar, but alternatively may be any pressure between about 1 E-2 and about 50 mbar.

[0046] In alternate embodiment methods, ion fragmentation is carried out in the ion trap. In further alternate embodiments, fragmentation of the precursor ions is carried out while the precursors are being accumulated in the ion trap or alternatively once a desired group of precursor ions has been accumulated. In a further alternate embodiment, the non-ergodic fragmentation method may be any known method including any photodissociation method - especially high energy photodissociation using photons of UV or shorter wavelengths - or any electron-based fragmentation method such as ETD, ECD, or EAD.

[0047] In a further alternate embodiment, the fragment ions are rapidly quenched following the fragmentation step. In a preferred embodiment, the fragment ions are cooled by collisions with inert gas in the rf ion trap in which the fragmentation step takes place. In a further alternate embodiment, at least some of the fragment ions are kinetically trapped in a conformation.

[0048] In a further alternate embodiment, the fragment ions need not be mass analyzed, but are only mobility analyzed in step e). In yet further alternate embodiments, the fragment ions are activated, preferably via energetic collisions with an inert gas (Clll), sufficiently to cause at least some of the fragment ions to unfold or change conformation. Alternatively, activation towards unfolding can be via any know ergodic method of activation including ion-molecule collisions, infrared activation, or excitation in a radio frequency field.

[0049] In another alternate embodiment, the fragment ions are analyzed according to any known prior art IMMS method including parallel accumulation serial fragmentation (PASEF), dia PASEF, midia PASEF, and / or dmf PASEF. According to a third preferred method, a) a hyphenated ion mobility mass spectrometer is provided, including at least one ion mobility analyzer and at least one mass analyzer wherein at least one of the ion mobility analyzers is preferably a TIMS analyzer and wherein at least one of the mass analyzers is preferably a TOF mass analyzer; b) gas phase analyte ions are formed from sample material by electrospray ionization; c) precursor ions of a known mobility are selected from the gas phase analyte ions, preferably via ion mobility analysis, most preferably via TIMS analysis, wherein precursors ions are allowed to pass an ion optical gate as they elute from the mobility analyzer whereas substantially all other ions are not; d) fragment ions are formed, non-ergodicly, from the selected precursor ions preferably via UVPD; e) the fragment ions are further activated preferably via energetic collisions with inert gas molecules; f) the resultant fragment ions are analyzed by ion mobility - mass spectrometry, preferably TIMS-TOF; and g) steps b) - f) are repeated a multitude of times, each time with a new group of ions of the same precursor, wherein the strength of the further activation in step e) is varied over a range of near zero, such that substantially no unfolding of the fragment ions is induced, up to an activation sufficient to induce unfolding of at least some of the fragment ions.

[0050] So, it is possible via a multitude of activation experiments to form Clll curves of the fragment ions. One can use the Clll curves to determine which of the fragment ions are representative of the native precursor conformation. One can use MS1 and MS2 experiments to identify the primary structure of each fragment ion - i.e. to determine what part of the protein (in terms of where in the amino acid sequence) the fragment came from Use the CIU curves to characterize the conformation of the fragment. One can further use the CIU curves to determine the original precursor conformation.

[0051] Generally (for all methods in this application), there can be a step c1) after step c) wherein fragment ions are immediately cooled to remove excess energy from the fragmentation process (so as to retain the fragment in its original conformation as long as possible). In alternate embodiments, the further activation may take the form of energetic collisions with an inert gas, collision with a surface, infrared photoactivation, or excitation in a radio frequency electric field.

[0052] In an alternate embodiment, said multitude of times in step g) is two times - one of which includes a near zero further activation and the other of which includes an activation sufficient to induce unfolding of at least some of the fragment ions. According to this embodiment, the results of step g) are differentially compared to one another and used to determine which fragment ions are substantially representative of the original precursor ions and / or to determine the native conformation of the precursor or otherwise limit its possible native conformations.

[0053] In alternate embodiments, said multitude of times in step g) is three or more times. According to this embodiment, the cross section, or equivalently mobility, of the fragment ions is analyzed as a function of the strength of the further activation. The results of the analysis are used to determine which fragment ions are substantially representative of the original precursor ions, and / or to determine the native conformation of the fragment ions or limit their possible native conformations, and / or to determine the native conformation of the precursor or otherwise limit its possible native conformations.

[0054] In alternate embodiments, any combination of one or more known ion mobility analyzer and one or more known mass analyzer may be used including drift cell, differential, travelling wave, or TIMS mobility analyzers, and quadrupole, ion cyclotron resonance, orbitrap, time- of-flight, or Paul trap mass analyzers.

[0055] In alternate embodiments, any known method of gas phase ion production including matrix assisted laser desorption ionization, atmospheric pressure chemical ionization, photo ionization, desorption electrospray ionization, or any other known method of ionization. The biopolymer may be any type of biopolymer, including proteins, peptides, oligonucleotides, or carbohydrates.

[0056] In alternate embodiments, the compounds in the sample material are separated from each other before the ionization step preferably using separation techniques known to maintain native conformations in biochemistry and those used in top-down analysis. This includes SEC, gel-based separations (like SDS-PAGE), CE, LC, and related approaches. In further alternate embodiments, the precursors are separated from other analyte ions and selected based on a combination of the precursors’ mass, mobility, and charge state.

[0057] In a further alternate embodiment, the precursor ions may be trapped and accumulated in an ion trap, preferably an rf ion trap. Preferably the ion trap is comprised of a set of conducting electrodes defining a volume in which ions the ions may be trapped. A radio frequency potential is applied to the electrodes which in turns produces a pseudopotential which confines the ions in the volume. Preferably the ion trap is filled with a gas of sufficient pressure to dampen the motion of the ions. The gas can be any combination of nitrogen, argon, helium, carbon dioxide, or any other inert gas. The static pressure of the gas is preferably about 3 mbar, but alternatively may be any pressure between about 1 E-2 and about 50 mbar.

[0058] In alternate embodiment methods, ion fragmentation is carried out in the ion trap. In further alternate embodiments, fragmentation of the precursor ions is carried out while the precursors are being accumulated in the ion trap or alternatively once a desired group of precursor ions has been accumulated. In a further alternate embodiment, the non-ergodic fragmentation method may be any known method including any photodissociation method - especially high energy photodissociation using photons of UV or shorter wavelengths - or any electron-based fragmentation method such as ETD, ECD, or EAD.

[0059] In a further alternate embodiment, the fragment ions are rapidly quenched following the fragmentation step. In a preferred embodiment, the fragment ions are cooled by collisions with inert gas in the rf ion trap in which the fragmentation step takes place. In a further alternate embodiment, at least some of the fragment ions are kinetically trapped in a conformation.

[0060] In a further alternate embodiment, the fragment ions need not be mass analyzed, but are only mobility analyzed in step f).

[0061] In another alternate embodiment the fragment ions are analyzed according to any known prior art IMMS method including parallel accumulation serial fragmentation (PASEF), dia PASEF, midia PASEF, and / or dmf PASEF. One key element of the present invention is that electrospray ionization is used to bring the biopolymer into the gas phase in an ionized state while retaining the biopolymer’s native structure, at least within the time span of these experiments.

[0062] In a first step this ionized, essentially native, (compact or extended) structure is selected and / or isolated, for example using IMS technologies.

[0063] The respective, still essentially native structure is then subjected to a fragmentation which retains the three-dimensional structure of the fragments as in the initial structure. To accomplish this, there is only very little distribution of the cleavage energy used for fragmentation to distort the fragment structures within the time span of the experiment, which is why the fragmentation is termed non-ergodic or essentially non-ergodic. This means that the fragmentation takes place essentially without disruption of the conformation of the fragment. This is in contrast to other fragmentation techniques, such as collision induced fragmentation, which leads to a redistribution of the collision energy substantially throughout the precursor and fragment ion, and a corresponding disruption of the three- dimensional structure of the fragments, leading to gas-phase annealed or extended fragments, which have lost the information of the initial native structure.

[0064] In the following step, these essentially native structure fragments are subjected to another second selection and / or isolation, for example using IMS technologies, and are then analyzed using mass spectrometry technologies.

[0065] There are two main strategies for determining which of the fragment ions are representative of the native conformation of the precursor:

[0066] 1) based on the differential analysis of native and non-native precursors (second preferred method described above); and

[0067] 2) based on the differential analysis of the fragment ions under strong vs weak activation towards fragment ion unfolding (third preferred method described above).

[0068] Either or both of these strategies can be used to evaluate the fragment ions.

[0069] In addition, the change in collision cross section, or equivalently mobility, with fragment ion activation - i.e. Clll - can provide further information as to the conformational structure of the fragment ion itself.

[0070] In accordance with the first strategy, precursor ions of a first, substantially native conformation of a target biopolymer are selected by mobility, charge state, and or mass; fragmented non-ergodicly; and then mass-mobility analyzed to identify the fragment ions (by MS1 and / or MS2 spectra) and their relative compactness. In a second step, precursors ions of a second, substantially non-native conformation of the same target biopolymer are selected by mobility, charge stage and / or mass; fragmented non-ergodicly; and then massmobility analyzed. If a compact fragment ion is produced from a native conformation precursor and a less compact fragment ion of the same type is produced from a non-native conformation of the same precursor type, then it can be concluded that compact fragment ion is more representative of the native conformation of the target biopolymer and the less compact fragment ion is more representative of the non-native conformation. The more compact fragment ion may then be used to infer something about the native structure of the target biopolymer.

[0071] In accordance with the second strategy, a first group of precursor ions of a substantially native conformation of a target biopolymer are selected by mobility, charge state, and or mass; fragmented non-ergodicly; and then mass-mobility analyzed to identify the fragment ions (by MS1 and / or MS2 spectra) and their relative compactness. In a second step, a second group of precursors ions of the same substantially native conformation of the same target biopolymer are selected by mobility, charge stage and / or mass; fragmented non- ergodicly; strongly activated toward unfolding (Clll); and then mass-mobility analyzed. If a compact fragment ion is produced from a native conformation precursor and a less compact fragment ion of the same type is produced after activation toward unfolding, then it can be concluded that compact fragment ion is not fully annealed and may be representative of the native conformation of the target biopolymer. The compact fragment ion may then be used to infer something about the native structure of the target biopolymer.

[0072] Alternatively and generally phrased for all the above three preferred methods, a method is proposed for the determination of the conformation of a biopolymer, preferably by using hyphenating ion mobility mass spectroscopy, preferably downstream of a chromatographic or electrophoretic separation.

[0073] The method includes the following steps a) ionization, in particular electrospray ionization (ESI), preferably downstream of said chromatographic or electrophoretic separation, is used for generating native or essentially native conformation ions of said of at least one biopolymer in the gas phase; b) these ions, or precursors based on these ions, are selected, preferably based on at least one of mobility, charge state, or a combination thereof, c) these selected ions or precursors are fragmented using a non-ergodic or essentially non- ergodic dissociation to form fragments; d1) the resulting fragments are on the one hand analyzed, normally without further activation, in particular without collision activation, for their distribution as a function of mass to charge, preferably in a survey spectrum, preferably using a selection based on at least one of mobility, charge state, or a combination thereof; d2) the resulting fragments are on the other hand analyzed, preferably in that they are either collision activated and then analyzed for their distribution: as a function of mobility to charge and / or as a function of mass to charge in at least one spectrum as a function of said collision activation, preferably analyzed using a selection based on at least one of mobility, charge state, or a combination thereof, or measured from further precursors (preferably in this case without collision activation) having a different conformation, e) using differential information from d1) and d2) about the fragments to determine which of the fragments resulting from step c) are in a native or essentially native conformation, and using information from the fragments having native conformation for the determination of the conformation of said at least one biopolymer.

[0074] One of the gists of the present invention as expressed broadly in step d1) and d2) and their differential analysis in e) as the comparison of the ion cross section (or mobility-to-charge) of an unactivated fragment ion to that of a collisionally activated (Clll) fragment ion or the cross section of an unactivated fragment ion from a precursor of a first collision cross section to that of an identical unactivated fragment ion (same sequence, same mass-to-charge) from an identical (same protein, same charge state) precursor of a second cross section. Regarding the above Liu et al reference it is noted that therein there is LIVPD and there is mention of CCS but no discussion of obtaining information related to higher order structure, i.e. conformational information. Conformations of fragments are mentioned and also how different conformations of fragments can be generated, but there is protocol how to determine conformational information in a differential way as claimed here. So there is no discussion of comparing mobilities of fragment ions as a function of collision activation or as a function of the cross section of the precursor. The reference is actually focused on determining the primary sequence of the target protein.

[0075] Regarding the above Fouque et al reference this was also focused on primary sequence including post translational modifications. Activation in all cases was focused on forming fragment ions (as opposed to unfolding the ions). Ion mobility was used to reduce interferences. Higher order structure was not discussed. Also, there is no disclosure of tandem TIMS operation.

[0076] One preferred experimental protocol is given as follows:

[0077] Step b) (preferably implemented by an ion mobility analyzer, preferably a trapped ion mobility analyzer, TIMS1) separates sample ions (presumably unfragmented protein ions) in the mobility domain. A group of (protein) ions of a selected type, based on mobility to charge is passed to a downstream trap (termed TO), collected and held there. The mobility to charge may represent not only the type of sample ion (e.g. protein) but also its conformation.

[0078] While in the downstream trap TO, these “precursor” ions are fragmented, e.g. irradiated (UV light) to induce fragmentation. Some of the fragments may retain secondary / tertiary structures representative of the precursor (protein) ions.

[0079] A fragment ion type (that is, having a particular molecular formula) is substantially distinguished from ions of other types, in the instrument, by its mass to charge ratio as measured in the (time of flight) mass analyzer. The mobility distribution of a fragment ion type is preferably measured via second ion mobility analyzer, preferably a trapped ion mobility analyzer, TIMS2 (providing the mobility to charge domain for the fragment ions) and a mass analyzer (e.g. TOF, providing the m / z domain). The “mobility distribution” for a particular fragment ion type may include ions of several different mobilities - each indicating a presumably different conformation.

[0080] According to an aspect of the present invention, at least two experiments are performed - the first experiment uses a first (presumably collisional) activation of the fragment ion type of interest (preferably no activation) and a second experiment uses a second activation of a second group of the same fragment ion type of interest.

[0081] And then the mobility distributions of the ion type of interest in the at least two experiments are compared to determine if the mobility distribution of the fragment ion type has changed due to activation. Such a change will potentially indicate “collision (or activation) induced unfolding”. Activation induced unfolding in turn will potentially indicate the unactivated mobility distribution is not fully unfolded but rather may include higher order structure resembling that of the original precursor ions.

[0082] Thus, the mass-to-charge as measured in the mass analyzer (e.g. TOF analyzer) is used to distinguish the fragment ion type of interest and the second ion mobility analyzer (TIMS2 analyzer) is used to follow the mobility-to-charge distribution of these ions as a function of activation.

[0083] An alternate method the “differential” comparison of the mobility distribution of the fragment ion type of interest is based on the mobility (conformation) of the selected precursor ions. In this case at least two experiments are performed - the first selecting precursors having a first mobility, and therefore a first conformation, in the first ion mobility analyzer (TIMS1) and a second experiment selecting precursors (presumed to be of the same mass to charge and same molecular formula as the first experiment) of a second mobility, and therefore a second conformation. The change in conformation of the precursor ions between the first and second experiments is expected to translate into changes in the conformations and therefore mobilities of at least some of the fragment ions produced e.g. by LIVPD in TO (no activation of the fragment ions is used or needed).

[0084] According to a preferred embodiment, the non-ergodic or essentially non-ergodic dissociation is selected from the group consisting of: ultraviolet dissociation (LIVPD) or electron dissociation (ED), in particular electron capture dissociation (ECD).

[0085] Preferably ultraviolet photo dissociation (LIVPD) is used with a laser, preferably at a frequency in the range of 150-350 nm, preferably in the range of 200-250 nm, and / or preferably for a time span of at least 10 ms, preferably of at least 20 ms or at least 40 ms, further preferably less than 1000 ms.

[0086] Selection can be carried out based on at least one of mobility, mass, charge state or a combination thereof.

[0087] Preferably selection is carried out by introducing ions into an ion mobility separator (IMS), preferably a trapped ion mobility spectrometry (TIMS) separator.

[0088] Preferably, selection is carried out by introducing ions into a trapped ion mobility spectrometry (TIMS) analyzer, wherein an ion trap is located between the two trapped ion mobility spectrometry (TIMS) analyzers, and wherein fragmentation is carried out while the respective ion is in the ion trap.

[0089] Collision activation can be carried out while the respective ion is located in an ion trap downstream of step b) or, preferably, can be carried out in the device used for selecting in step d2). Preferably the said device used for selecting is a trapped ion mobility spectrometry (TIMS) analyzer.

[0090] According to yet another preferred embodiment, the device used for selection is a trapped ion mobility spectrometry (TIMS) analyzer, and at least in an accumulation section (11a) thereof collision induced unfolding can be carried out by gas collisions, preferably by applying voltages of at least 5 V or at least 50 V or in the range of 100-200 V.

[0091] Fragments with native or essentially native conformation can be identified by the fact that the respective fragments are reducing or disappearing from the spectrum as a function of collision activation, and for the determination of the conformation of the at least one biopolymer, the collision cross-section measured of that fragment can be used, preferably assisted by providing database information and / or molecular simulation information, in particular based on the primary structure of that fragment.

[0092] Preferably, the biopolymer is selected from the group of proteins, peptides, nucleotides, polysaccharides, natural rubbers, lignin, suberin, cutin, cutan, melanin, polyhydroxyalkanoates as well as mixtures and derivatives thereof, wherein preferably as a starting material for the biopolymer conformation determination also a mixture can be used, including proteomics samples such as plasma and urine samples.

[0093] Particularly preferably, the biopolymer is a protein, and the conformation to be elucidated is the secondary and / or tertiary and / or quaternary structure thereof.

[0094] In at least one step, ion selection can be carried out by introducing ions into a trapped ion mobility spectrometry (TIMS) separator, wherein said trapped ion mobility spectrometry (TIMS) separator is a TIMS analyzer with parallel accumulation and separation, operating using a method comprising the steps:

[0095] (a) accumulating ions entering in an RF ion trap;

[0096] (b) transferring at least a subset of the accumulated ions into a trapping ion mobility separator, in which the transferred ions are radially confined by an RF field and pushed by a gas flow against a rising edge of an axial electric DC field barrier such that the transferred ions are spatially separated along the rising edge according to ion mobility;

[0097] (c) successively releasing the transferred ions according to their ion mobility by decreasing the height of the electric DC field barrier while ions from the ion source are further accumulated in the RF ion trap; and

[0098] (d) restoring the height of the electric DC field barrier which triggers a consecutive transfer of the accumulated ions from the RF ion trap into the trapping ion mobility separator.

[0099] According to yet another preferred embodiment, at least one or in both of steps b) and d), selection is carried out by introducing ions into a trapped ion mobility spectrometry (TIMS) separator, wherein said trapped ion mobility spectrometry (TIMS) separator is a TIMS analyzer, and wherein separating ions in time according to mobility in the TIMS analyzer involves driving the ions transferred to the TIMS analyzer by a first gas flow against a first counter-acting electric DC field barrier such that the ions are trapped and spatially separated according to their mobilities at different positions along a ramp of the first electric DC field barrier at which a friction force of the first gas flow equals the counter-acting force of the first electric DC field barrier, and temporally separating ions according to mobility in the TIMS analyzer by adjusting a height of the first electric DC field barrier or the velocity of the first gas flow.

[0100] In addition to this method, the present invention relates to an apparatus. According to a preferred embodiment, this is an apparatus for the determination of the conformation of at least one biopolymer by using hyphenating mass spectrometry with electrophoretic or chromatographic separation, in particular for carrying out the above method, said apparatus preferably comprising: if needed means for electrophoretic or chromatographic separation;

[0101] (if present downstream of the chromatographic separation) electrospray ionization (ESI) for generating native or essentially native conformation ions of said of at least one biopolymer in the gas phase; a first ion mobility separator (IMS) for receiving and sequentially releasing said native or essentially native conformation ions of said of at least one biopolymer ions from said IMS according to their ion mobility, an ion trap downstream of said ion mobility separator (IMS) and means for fragmenting using a non-ergodic or essentially non-ergodic dissociation to form fragments; a second ion mobility separator (IMS) for receiving and sequentially releasing said fragments from said IMS according to their ion mobility and allowing for collision activation, an apparatus for carrying out a mass spectrometry measurement on said fragment ions. The first and second ion mobility separator can be a TIMS analyzer, preferably a TIMS analyzer with parallel accumulation and separation, in particular operating using a method comprising the steps:

[0102] (a) accumulating ions in an RF ion trap;

[0103] (b) transferring at least a subset of the accumulated ions into a trapping ion mobility separator, in which the transferred ions are radially confined by an RF field and pushed by a gas flow against a rising edge of an axial electric DC field barrier such that the transferred ions are spatially separated along the rising edge according to ion mobility;

[0104] (c) successively releasing the transferred ions according to their ion mobility by decreasing the height of the electric DC field barrier while ions from the ion source are further accumulated in the RF ion trap; and (d) restoring the height of the electric DC field barrier which triggers a consecutive transfer of the accumulated ions from the RF ion trap into the trapping ion mobility separator.

[0105] According to a preferred embodiment, the main axis of the first ion mobility separator is perpendicular to the main axis of the second ion mobility separator, and wherein a UV laser is provided for irradiating the ion trap for non-ergodic or essentially non-ergodic dissociation. The first ion mobility separator can be a TIMS analyzer, in which the ions are radially confined by an RF field and pushed by a gas flow against a rising edge of an axial electric DC field barrier such that the transferred ions are spatially separated along the rising edge according to ion mobility, and wherein the height of the electric DC field barrier is chosen such as to retain only ions having an ion mobility higher than said upper threshold value.

[0106] The second ion mobility separator can be a TIMS analyzer with parallel accumulation and separation, in which the transferred ions are accumulated in an accumulation section from the first ion mobility separator in an RF ion trap, in which the ions are radially confined by an RF field and pushed by a gas flow against a rising edge of an axial electric DC field barrier such that the transferred ions are spatially separated along the rising edge according to ion mobility, and wherein the height of the electric DC field barrier is chosen such as to retain only ions having an ion mobility higher than said lower threshold value followed by transferring at least a subset of the accumulated ions into a trapping ion mobility separator in a separation section, radially confined by an RF field and pushed by a gas flow against a rising edge of an axial electric DC field barrier such that the transferred ions are spatially separated along the rising edge according to ion mobility, and wherein the height of the electric DC field barrier at the beginning of the release is preferably chosen such as to retain only ions having an ion mobility higher than said lower threshold value, wherein preferably between the accumulation section and the separation section of the second ion mobility separator a second ion gate is provided preventing ions having an ion mobility higher than said lower threshold value to enter the separation section.

[0107] Further embodiments of the invention are laid down in the dependent claims.

[0108] BRIEF DESCRIPTION OF THE DRAWINGS

[0109] Preferred embodiments of the invention are described in the following with reference to the drawings, which are for the purpose of illustrating the present preferred embodiments of the invention and not for the purpose of limiting the same. In the drawings,

[0110] Fig. 1 schematically shows the principle as proposed in the prior art document US 2012 / 0273673 for sequential field barrier filtering in TIMS-like devices;

[0111] Fig. 2 shows the device and operation of a trapped ion mobility spectrometry (TIMS) analyzer. Top: Scheme of the TIMS device. Center: The profile of the electric field strength along the z axis. The size of the dots on the electric field ramp reflects their space charge. The scan releases the ion bunches in sequence of their mobilities. Bottom: The ion current of ion pulses separated in time according to mobility, representing an ion mobility spectrum.

[0112] Fig. 3 shows the device and operation of a trapped ion mobility spectrometry (TIMS) analyzer with parallel accumulation and separation. Top: Scheme of the TIMS device depicting “accumulation” (11a) and “analyzer” (11 b) regions. Bottom: Plots representing the DC electric field gradient and ion positions during the accumulation and analysis steps. In the upper row a first group of ions are eluted from the analyzer region 11b by lowering the electric field gradient between points 32 and 34 while a second group of ions is simultaneously accumulated in the accumulation region on the electric field gradient between points 30 and 31. Note the size of the dots represents the collision cross section of the ions. In the lower row, the field in the analyzer region 11 b is returned to its full strength after having eluted the first group of ions and the second group of ions is transferred from the accumulation region 11a to the analyzer region 11b by dropping the DC field strength in the accumulation region 11a to zero.

[0113] Fig. 4 shows a schematic overview of a mass spectrometer with two TIMS analyzers and an ion gate in between.

[0114] Fig. 5 shows schematically a device and operation of the tandem TIMS device of Fig. 4, with a first TIMS analyzer (TIMS 1), an ion gate, and a second TIMS analyzer (TIMS 2). The diagram below shows ions collected on the field ramp of TIMS 1 , with a marked range of ion mobility representing the ions of interest. The size of the dots represents the amount of ions of each type, thereby indicating the space charge of these ions. During the TIMS 1 scan the ions of interest are selected by the gate and collected on the electric field ramp of TIMS 2. The selected ions can be analyzed via TIMS 2 without further space charge disturbance.

[0115] Fig. 6 shows in a) a schematic overview over the proposed process for the determination of native -like protein fragments, in b) shows a schematic of the three-dimensional structure of ubiquitin, and in c) shows a schematic overview over a device suitable and adapted to carry out the proposed protocol;

[0116] Fig. 7 shows another schematic overview over a device suitable and adapted to carry out the proposed protocol with further detail;

[0117] Fig. 8 on the left side schematically shows the situation where behind TIMS1 the native molecules are subjected to collisional induced dissociation (CID) and the annealed fragments are then analyzed in TIMS2, and on the right side schematically the situation where behind TIMS1 ergodic dissociation (in this case LIVPD) is carried out and the essentially native fragments are analyzed in TIMS2, if need be under differentiating collisional activation conditions;

[0118] Fig. 9 shows UV photodissociation (LIVPD) of the folded, native-like ubiquitin conformation produces compact, folded fragment ions from cleavage at Pro19 and Pro37. (A) ESI mass spectrum for ubiquitin showing charge states 7+ and 8+. (B) The ion mobility spectrum of charge state 8+ shows a predominant compact, native-like conformation (1283 A2, blue) and a minor feature (1947 A2, gray) of extended, gas-phase unfolded structures. (C) Ion mobility spectra recorded for fragment ions at m / z 1001 (b36), m / z 1141 (y40), and m / z 1306 (y58) produced from the compact, native-like ubiquitin conformation by LIVPD at the Pro19 and Pro37 residues (blue traces) differ strongly from those produced from the extended, gas-phase unfolded ubiquitin conformation (gray traces). (D) Ion mobility spectra recorded for the corresponding fragment ions produced by collision-induced dissociation (CID) of the compact, native-like ubiquitin conformation (blue traces) are similar to those produced from the extended, gas-phase unfolded ubiquitin conformation by CID (gray traces) or LIVPD (C, gray traces);. in a) the mass spectrum for ubiquitin for the charge states 8+ and 7+, in b) shows the ion mobility spectrum for charge state 8+, in c) shows the spectra for the compact precursor and the extended precursor for different ions for LIVPD conditions, and in d) shows these ions under CID conditions;

[0119] Fig. 10 shows a comparison of the fragment ions detected following collision-induced dissociation (CID) and UV photodissociation (UVPD) of ubiquitin conformations, (a) Mass spectrum recorded after mobility-selecting the compact conformation of ubiquitin charge state 8+, [M+8H]8+. (b) Mass spectrum recorded after performing CID in the TIMS-1 / TIMS-2 interface of the selected compact conformation of ubiquitin charge state 8+. The spectrum shows the presence of an, bn, yn - type ions that are typically observed for CID of a polypeptide chain, (c), (d) Mass spectra recorded following UV photodissociation (UVPD) of the compact (c) and extended (d) conformations of ubiquitin charge state 8+ The spectra are void of the traditional an, bn, yn - type fragment ions seen under CID in (b) but show the intense formation of [a+1], [a+2], [b+2], [y-2] fragment ions that are typical for UV photodissociation of proline-containing polypeptides. The observation that the fragment ions from CID are not prominently observed upon UV dissociation indicates that a vibrationally hot electronic ground state of ubiquitin does not play a significant role in forming the fragment ions in (c), (d).; Fig. 11 shows a comparison of y585+ and [y58-2]5+ fragment ions isotope patterns to the experimental abundances under various measurement conditions, (a) CID in the TIMS-1 / TIMS-2 interface produces exclusively y585+ ions, (b), (c) UV photodissociation of the compact and extended conformations produces exclusively [y58-2]5+ fragment ions, (d), (e) The y585+ and [y58-2]5+ fragment ions are not observed if the ions are not irradiated by UV photons or without storage in the ion trap, (f), (g) Fragmentation of the [y58-2]5+ fragment ion is not observed upon collision-activation in TIMS-2.;

[0120] Fig. 12 shows abundances as a function of the collision cross sections for variable trap and UVPD irradiation times, compared with CID conditions;

[0121] Fig. 13 shows the cross-sections calculated for the fragment ion models in a) and the structures thereof in b);

[0122] Fig. 14 shows the distribution as a function of the collision cross-section for the compact precursor without vibrational activation, for the compact precursor under vibrational activation with 160 V, and for the extended precursor, wherein in a) the situation for UVPD and in b) the situation for CID are shown;

[0123] Fig. 15 shows in a) the normalized abundance as a function of the intact protein collision cross-section, in b) the collision cross-section as a function of the intact protein collision cross-section, in c) the average structure based on intact protein and fragment ions and in d) the average structure of the intact protein based on collision cross-section.

[0124] Fig. 16 shows (A) Collision cross-section distribution of an MD-derived set of native-like (green) and gas-phase (yellow) ubiquitin models demonstrating that the nativelike and gas-phase models differ in the collision cross sections of their fragment ions. (B) Cartoon illustrating a potential computational approach to propose complete structures of biopolymers from the measured cross sections of the intact biopolymer and its fragments. (C) selection of mainly native models considering the cross-sections of the intact polypeptide and the fragment ions. (D) the ability of supervised learning / machine learning methods selects predominantly native-like biopolymer models from a large set of incorrect models.

[0125] DESCRIPTION OF PREFERRED EMBODIMENTS

[0126] Fig. 2 outlines schematically a TIMS analyzer which can be used in the present context as referred to in US 7,838,826 B1 or in US 10,241 ,079 B1 and its operation. Entrained by a gas 7, ions 6 from an electrospray ion source (not shown) are introduced via capillary 8 into a first chamber of a vacuum system. A repeller plate 9 drives the ions 6 into an entrance funnel 10 of the mobility analyzer. Ion funnels 10, 12 usually are built as a stack of apertured diaphragms the openings of which taper to smaller diameters thus forming an inner volume in the shape of a funnel. Two phases of an RF voltage are applied alternately to the diaphragms to build up a pseudopotential which keeps the ions away from the funnel walls. The ions are driven to and through the narrow end of the first funnel 10 into the TIMS tube 11 by an axial gas flow 14 and optionally by an additional DC potential gradient along the diaphragms.

[0127] The axial gas flow 14 through the TIMS tube 11 is laminar and shows, in radial direction, a substantially parabolic velocity distribution. Nitrogen may serve as a preferred gas. The vacuum conditions around the TIMS tube 11 are chosen such that the maximum gas velocity amounts to about 100 to 150 meters per second, at a pressure of a few hectopascals. This velocity is only achieved near the axis. Further off axis, the velocity is considerably smaller, as indicated by the arrows 14 in Fig. 2.

[0128] The first funnel 10 guides the ions into the TIMS tube 11 forming a tunnel with internal RF quadrupole field in radial direction. The TIMS tunnel 11 comprises a stack of thin electrodes with central holes which form a circular tube arranged around the z-axis of the device. The thin electrodes are separated by insulating material closing the gaps between the electrodes around the tube. The electrodes of the TIMS tube 11 are segmented into quadrants 1 , 2, 3, 4, to allow for the generation of a radially confining quadrupolar electric RF field inside. The quadrants 1 , 2, 3, 4 of the tube electrodes are shown at the top of Fig. 2 with equipotential lines of the quadrupolar RF field inside the tube at a given time. It should be mentioned here that the design of a quadrupole tunnel does not necessarily consist of metal electrode sheets; there are a lot of different possibilities including stacked PCB boards or even a rolled PCB board with printed electrodes.

[0129] Inside the TIMS tunnel 11 , the ions are blown by the gas flow 14 against an axial electric DC field barrier. In the center part of Fig. 2, the profile of the axial electric DC field barrier is shown for three phases of a scan. Between z locations 20 and 23, the electric DC field increases linearly, generated by a quadratically increasing electric potential. Between z locations 23 and 24, the electric DC field remains constant, forming a plateau of the electric DC field barrier, generated by a linear increase of the electrical potential. In a simple device, for instance, the complete field profile can be generated by a single voltage, applied to the diaphragm electrode at location 24, and divided by precision resistors along the diaphragm electrodes of the TIMS tube 11. The resistors between location 20) and 23 increase linearly, the resistors between 23 and 24 have equal resistance. In more complex devices, nonlinear field electric field profiles may be generated, even adjustable DC field profiles, e.g. by digital-to-analog converters (DAC).

[0130] The conventional operation of the TIMS analyzer starts with an “ion accumulation phase”, accumulating ions on the uppermost electric DC field ramp of the diagram. A voltage difference on the order of 300 volt produces the electric DC field barrier. The ions are blown by the gas flow, symbolically indicated by the arrows 16, against the electric DC field barrier and are stopped there because they cannot surmount the electric DC field barrier. It should be noted that the arrows 16 represent the maximum gas velocity of the parabolic gas velocity distribution 14 within the tube. The ions are accumulated on the rising edge of the electric DC field between locations 20 and 23, where ions of low mobility (mainly heavy ions with large collision cross section) gather in the high field near the upper end of the field ramp, whereas ions of high mobility gather in the low field near the foot of the ramp. The size of the dots represents the abundance of the ions of distinct ion mobility, indicating the strength of the space charge. In the subsequent “scan phase”, the supply voltage for the electric DC field barrier is steadily decreased, and ions of increasing mobility can escape towards an ion detector, particularly to a mass analyzer operating as ion detector. In the bottom of the figure, the resulting ion current of the released ion species is shown. The measured total ion current curve i= f(t) presents directly an ion mobility spectrum from low ion mobilities to high ion mobilities.

[0131] The ion mobility resolution RmOb depends on the scan speed. The lower the scan speed, the higher the resolution. As already mentioned, ion mobilities of RmOb = 400 have been achieved with the comparably small devices, using slow scans. Since the ions generated in the ion source are lost during the scan phases, the duty cycle (or the utilization rate of the ions) depends on the ratio of the accumulation time ta to the scan time ts.

[0132] Fig. 3 shows an embodiment of a mobility spectrometer which can be used in the present context, and which combines accumulation and subsequent separation as e.g. described in US 9,683,964 B2. The TIMS comprises an elongated tunnel 11 , divided into an accumulation unit 11a and a scan unit 11b, and two voltage supply units (not shown) for the two tunnel units 11a, 11b, contacting the diaphragms at locations 31 and 34. Chains of resistors between the diaphragms in both tunnel units produce two axial electric DC field profiles, shown in the bottom part of the figure.

[0133] The operation of the TIMS device according to Fig. 3 comprises two phases: In the accumulation and scan phase, ions from an ion source (not shown) are accumulated on the rising edge of the electric field profile in the accumulation unit 11a while, at the same time, ions in the scan unit 11 b are scanned by decreasing the voltage supplied to location 34 of the scan unit 11b, thereby releasing ions with higher and higher mobilities through the exit funnel 13 towards the ion detector.

[0134] In the transfer phase, first the voltage of the scan unit 11b is restored, and then the voltage of the accumulation unit 11a is switched off to let the ions be driven by the gas flow onto the rising edge of the electric field profile of the scan unit 11b. The transfer is completed after only one millisecond, and the accumulation and scan phase may start again by switching on the voltage at location 31 .

[0135] Fig. 4 shows a time-of-flight mass spectrometer with two TIMS analyzers in a tandem arrangement as e.g. described in US-A-2018340910. In the present context, this is made use of in that the first ion mobility analyzer (TIMS 1 , which can be of the type as described in Fig. 2 or in Fig. 3) may scan a bunch of collected ions, thereby releasing ions in sequence of their ion mobilities. During the scan, the ion gate is alternately opened and closed, closed to reflect (or neutralize) unwanted ions and opened to pass ions in distinct ranges of mobility. In this way, highly abundant ions with their space charge can be reflected in full or at least partially. The passing ions enter the second ion mobility analyzer (TIMS 2, which can be of the type as described in Fig. 2 or in Fig. 3, but is preferably of the type of Fig. 3 in the present context) where they can be analyzed according to their ion mobility with high ion mobility resolution, undisturbed by space charge. The mass spectrometer then can measure their precise masses; a tandem mass spectrometer with quadrupole mass filter and time-of-flight analyzer, as presented in Fig. 3, even can measure fragment ion spectra for a better identification of the ion species.

[0136] The ions transferred to TIMS 2 may stem from a single range of mobility, or from several ranges, selected by switching the gate accordingly. In one embodiment, ions from a single range of mobility are collected on a flat ramp of TIMS 2, to spread the ions as far as possible along the z axis of TIMS 2. In this mode, ions may be accumulated and scanned in TIMS 1 several times to accumulate as many ions in TIMS 2 as required for an analysis of high quality. If there is a kind of ion with extremely high abundance within the range of interest, only a small portion of these ions may be transferred by reflecting the largest part of these ions. The ion gate preferably switches faster than the temporal width of ion pulses leaving the first ion mobility analyzer. The length of an ion pulse released by the scan is in the order of a millisecond, whereas the switching time for the gate can be below a microsecond; the ion pulse therefore can easily be cut into portions.

[0137] Whereas the radially confining RF field of TIMS 2 is preferably quadrupolar in order to achieve a high ion mobility resolution, TIMS 1 may show a tube with higher inner diameter, and / or with radial RF fields of higher multitude, like hexapole, octopole, or dodecapole, or with an RF tunnel. An ion trap can additionally be located upstream of TIMS 1 to accumulate ions from the ion source during the scan of TIMS 1. If the trap can be mass selectively unloaded, even TIMS 1 may be relieved from space charge.

[0138] Fig. 5 illustrates, in the top part, two tandem TIMS devices (TIMS 1 , TIMS 2) separated by an ion gate. The ion gate is formed as an ion-optical einzel lens. The gate can be switched on and off in less than a microsecond. During the scan of TIMS 1 , the gate may be opened and closed in a suitable manner to pick out ions of interest. The ions of interest transferred to TIMS 2 may stem from a single range of mobility, as shown in the bottom part of Fig. 5, which show the scheme for the situation where for TIMS1 and for TIMS 2 the type as described in Fig. 2 is used.

[0139] Fig. 6 shows in a) a schematic overview over the proposed process for the determination of native -like protein fragments, in b) shows a schematic of the three-dimensional structure of ubiquitin, and in c) shows a schematic overview over a device suitable and adapted to carry out the proposed protocol.

[0140] Fig. 7 shows a schematic overview over such a device suitable and adapted to carry out the proposed protocol, wherein further options are indicated, in particular the possibility to have collision induced dissociation (CID), or collision induced unfolding (Clll) downstream of TIMS-1 at the aperture plate L1 , how collision induced unfolding (Clll) is forced by applying a high voltage and gas molecule collisions in TIMS-2, and how it is possible to further have surface induced dissociation (SID) downstream of the quadrupole.

[0141] Fig. 8 shows schematically what is happening under collision induced dissociation (CID) conditions between the two TIMS devices (left) and what is happening under LIVPD (right). In the latter case fragments enter the second TIMS device in compact form, and if desired, the cross-sectional properties thereof are measured in that TIMS2 device or they are subjected to collisional activation to differentiate between fragments which were already annealed before entering the TIMS2 device and those which were in the compact form upon entry.

[0142] To investigate if irradiation of an intact solvent-free protein with UV photons produces protein fragments without significantly distorting their native structures, we selected the 76- residue a / p protein bovine ubiquitin (8.6 kDa). The tertiary structure of ubiquitin has been extensively characterized, comprises a p-sheet with five antiparallel p-strands and a single helical segment, and exhibits limited structural flexibility in its native state (Fig. 6b). Previous ion mobility / ion mobility / mass spectrometry experiments showed that folded, solvent-free ubiquitin ions exist in a predominantly native fold with an intact p-grasp motif and a-helix. Energetic activation or prolonged storage of the native-like ubiquitin ions in the solvent-free environment promotes their denaturation into gas-phase structures. We reasoned that comparing the structures of fragment ions produced from dissociating native-like folded and gas-phase unfolded ubiquitin conformations of the same charge state would reveal if fragment ions produced from backbone cleavage of ubiquitin retain aspects of their native structures.

[0143] The instrumental basis of our approach is our recently developed tandem-trapped ion mobility spectrometer coupled with an ion trap, a laser for irradiation of the trapped ions with UV photons, and a quadrupole time-of-flight mass spectrometer (Fig. 6c as well as Fig. 7 and Fig. 8) for mass analysis. Protein ions are generated using native electrospray ionization and are gently transferred through the instrument components. Tandem-TIMS first separates distinct protein conformations and measures their collisional cross sections in TIMS-1 by the times required to traverse neutral gas molecules against an applied weak electric field. Electrodes downstream of TIMS-1 are used to select the desired protein conformation after elution from TIMS-1 , followed by storing the ions in the trap and irradiating them with UV photons leading to dissociation (UVPD). If the stored proteins dissociate into fragments, their collision cross-sections are subsequently measured in TIMS-2, and their mass-to-charge ratios are determined. As illustrated in Fig. 6, this instrument can record multiple collisional cross-sections of an intact protein that could potentially be useful to characterize the overall protein shape (intact protein, TIMS-1) and its internal structure (protein fragment ions, TIMS-2).

[0144] Charge states 7+ and 8+ predominate the mass spectrum recorded for ubiquitin in positive electrospray ionization from an aqueous solution at pH 3.5 without fragmentation (Fig. 9a), where its native state prevails. These charge states are dominated by compact ubiquitin conformations with collision cross sections of 1242 A2and 1283 A2for 7+ and 8+, respectively (Fig. 9b), with charge state 8+ showing an additional minor feature at -1947 A2. Previous time-resolved tandem-TIMS experiments showed that the compact features retain >80% of their native contacts under typical IM / MS measurements, whereas the extended feature corresponds to a gas-phase structure due to the unfolding of the compact conformation in the solvent-free environment. The spectra for charge states 7+ and 8+ thus indicate that the compact features reflect essentially native conformations, whereas the extended conformation of charge state 8+ reflects a gas-phase structure (also termed annealed structure).

[0145] We thus reasoned that if the dissociated protein fragments were to retain the structure and non-covalent contacts of the intact protein, the compact and extended ubiquitin conformations would dissociate into fragments with substantially different conformations. To test this hypothesis, we selected the compact and extended ubiquitin ions of charge state 8+ after elution from TIMS-1 , dissociated the backbone of the selected ubiquitin ions by irradiation with UV photons (UVPD), and recorded the collision cross sections and mass- to-charge ratios of the produced fragment ions (Fig. 9c and Fig. 10). The recorded mass spectra confirm that irradiation with UV photons dissociates the ubiquitin backbone and produces sequence-informative fragment ions, see Fig. 10, which shows a comparison of the fragment ions detected following collision-induced dissociation (CID) and UV photodissociation (UVPD) of ubiquitin conformations, (a) shows a mass spectrum recorded after mobility-selecting the compact conformation of ubiquitin charge state 8+, [M+8H]8+.

[0146] (b) shows a mass spectrum recorded after performing CID in the TIMS-1 / TIMS-2 interface of the selected compact conformation of ubiquitin charge state 8+. The spectrum shows the presence of an, bn, yn - type ions that are typically observed for CID of a polypeptide chain.

[0147] (c), (d) show mass spectra recorded following UV photodissociation (UVPD) of the compact

[0148] (c) and extended (d) conformations of ubiquitin charge state 8+. The spectra are void of the traditional an, bn, yn - type fragment ions seen under CID in (b) but show the intense formation of [a+1], [a+2], [b+2], [y-2] fragment ions that are typical for UV photodissociation of proline-containing polypeptides. The observation that the fragment ions from CID are not prominently observed upon UV dissociation indicates that a vibrationally hot electronic ground state of ubiquitin does not play a significant role in forming the fragment ions in (c),

[0149] (d).

[0150] The dominant UV-induced fragment ions are [£>36+2]4+and [y4o-2]4+from cleavage at Pro37 and [ys8-2]5+from cleavage at Pro19. The observation of [b+2] and [y-2] fragment ions indicates the abstraction of two hydrogen atoms during amide bond cleavage, a hallmark of a radical-based dissociation mechanism associated with UV-induced photodissociation of polypeptide backbones; the dominance of cleavage at the Pro19 and Pro37 residues is consistent with previous findings. We next verified that these UV-induced fragment ions are observed when either the compact or the extended ubiquitin conformations are stored in the trap and irradiated with UV photons (Fig. 11 b,c). We confirmed that they are not detected without irradiation by UV photons (Fig. 11a,d,g) or when UV photons irradiate the ions without storage in the ion trap (Fig. 11e).

[0151] We further validated, by varying the storage time in the ion trap from 0 ms to 500 ms, that the intensity of these radical-induced fragment ions increases with increasing exposure time to UV photons (Fig. 12). Next, we corroborated that collisional activation between the accumulation and separation regions of TIMS-2 and downstream of TIMS-2 does not produce amide bond cleavage using the applied instrument settings. Finally, we confirmed that collisional-induced dissociation (CID) of ubiquitin in the TIMS-1 / TIMS-2 interface does not produce the radical-based [b36+2]4+, [y4o-2]4+or [ys8-2]5+ions but, instead, their £>364+, y4o4+or ys85+congeners (Fig. 11a). Hence, our data underline that the radical-based fragment ions [£>36+2]4+, [y4o-2]4+, and [ys8-2]5+are exclusively detected if ubiquitin ions are irradiated with UV photons while being stored in the ion trap.

[0152] These observations underline that a vibrationally hot electronic ground state of the protein does not produce the [bn+2] and [yn-2] fragment ions. Taken together, our data thus underscore that the [bn+2] and [yn-2] fragment ions related to cleavage at Pro19 and Pro37 are most likely produced via an electronically excited protein state induced by the absorption of 213 nm photons.

[0153] Fig. 9c confirms our hypothesis that UV photodissociation of the folded, native-like ubiquitin conformation produces fragment ions that retain a folded conformation. Specifically, the figure shows that the structures of the UV-induced fragment ions [£>36+2]4+, [y4o-2]4+, and [ys8-2]5+correlate with the ubiquitin conformations of their intact precursors. Specifically, when the folded, native-like ubiquitin conformation is UV photodissociated, then the collision cross-sections recorded in TIMS-2 for the [£>36+2]4+, [y4o-2]4+, and [ys8-2]5+fragment ions are roughly 20% smaller than when the extended, gas-phase conformation is dissociated. Hence, our data show that UV photodissociation of the compact, folded ubiquitin conformation produces compact, folded fragment ions. By contrast, UV photodissociation of the extended, gas-phase ubiquitin conformation produces extended, unfolded fragment ions.

[0154] As a negative control, we produced fragment ions from the compact and extended ubiquitin conformations through slow vibrational activation using collisional-induced dissociation (CID, see Fig. 9d). This dissociation method is known to produce protein fragment ions that are uncorrelated to the intact protein structure. In line with these expectations, the fragment ions produced by CID exhibit only minor structural differences in terms of the number of features in the spectra, their collision cross sections, or relative abundances irrespective of the initially selected ubiquitin conformations. Indeed, the main features of the CID fragment ion spectra are within 1% to 5% of those produced by UV-irradiation of the extended gasphase ubiquitin conformation in terms of their collision cross sections.

[0155] Our data thus show that

[0156] (1) the shape of the UV-induced fragment ions correlates with that of the precursor ion: a compact precursor ion correlates with a compact fragment ion after UV-irradiation, and vice versa,

[0157] (2) this correlation between the intact protein structure and the fragment ion structure is a property exclusive to the UV-induced fragment ions. It follows, therefore, that amide bond cleavage upon UV irradiation under the conditions employed in our tandem-TIMS instrument occurs on a faster timescale than the structural changes of the polypeptide backbone. Consequently, the most likely interpretation of our data is that the fragment ions produced by UV irradiation retain aspects of their native-like precursor ubiquitin structure. The proposition that fragment ions exist in a folded, native-like conformation and have not yet isomerized into an energetically favorable gas-phase conformation can be experimentally tested by examining their collision cross sections as a function of their vibrational activation (collision-induced unfolding / CIU): If the fragment ions are indeed produced in a folded, native-like structure, they should isomerize into their gas-phase conformations upon vibrational activation. Specifically, it is expected that the gas-phase isomerized fragment ions would adopt structures resembling those produced by CID (Fig. 9). We thus vibrationally activated the compact fragment ions by energetic collision with buffer gas molecules, which we accomplished by increasing the electric potential between electrodes T2 and T3 in TIMS-2 as described. Fig. 14A shows that the folded, compact [y4o- 2]4+and [yss-2]5+fragment ions unfold into structures with approximately 20% larger collision cross-sections when the electric potential is raised from 5 V to 160 V. By contrast, collisional activation of the fragment ions produced by CID did not result in detectable structural changes (Fig. 14B). Indeed, the collision cross-sections of the vibrationally-activated [ / 4o- 2]4+and [yss-2]5+fragment ions (965 A2and 1265 A2) closely match the cross sections observed for the fragments produced from UV photodissociation of the extended, gasphase ubiquitin conformation (980 A2and 1270 A2) and those of the gas-phase isomerized fragment ions produced by CID (955 A2and 1285 A2, respectively; Figs. 9 and 14). Taken together, these results show that UV photodissociation of the native-like ubiquitin conformation produces folded fragment ions that require energetic activation to isomerize into energetically favorable gas-phase conformations. This supports the view that the folded fragment ions retain aspects of their compact, native-like ubiquitin precursor conformation. To identify what aspects of the native ubiquitin structures are most likely preserved in the folded protein fragments produced by UV photodissociation, we generated model structures for the fragment ions using molecular dynamics simulations. Building on the structural ensemble computed by our structure relaxation approximation (SRA) for the compact, native-like ubiquitin conformation of charge state 8+, we generated model structures for the / 4o and yss fragment ions and calculated their collision cross-sections. Fig. 13A shows that the collision cross-sections calculated for the fragment ion model structures match the experimental cross-sections of the folded UV-induced fragments to better than 5% (-825 A2and -880 A2versus -855 A2for y4o; -1060 A2versus -1010 A2for yss). Next, we simulated the vibrational activation of the fragment ions using molecular dynamics at increasing simulation temperatures. Fig. 13A shows that the simulated vibrational heating of the fragment ions approximates the experimental cross-section differences of -20% between the vibrationally non-activated and activated fragment ions. Given the consistency between the experimental and computed fragment ion spectra, we used the computed models to interpret the structures of the folded fragment ions observed in the experiment. Comparison of the model structures to the equivalent polypeptide fragments in the ubiquitin x-ray structure (PDB 1 LIBQ) suggests that the folded, compact [ / 4o-2]4+and [yss-2]5+fragment ions largely retain their native p-sheet and a-helix components (Fig. 13B). Overall, the models exhibit a mean fraction of native contacts Q of approximately 0.75 and a mean Caroot-mean-square deviation of ~3.5 A, with native contacts mainly lost at the flexible loops / turns and the fragment termini. We rationalize the proposed significant retention of the p-sheet and a-helix after UV photodissociation because the proline residues are located in the flexible regions of the polypeptide chain between these a / p-regions. Hence, cleavage at the prolines would only minimally perturb hydrogen-bond networks within the secondary structure elements. Taken together, our data strongly support the notion that UV photodissociation of the compact, folded ubiquitin conformation produces protein fragments with native-like structures.

[0158] Taken together, our data show that native-like structures can be preserved in fragment ions produced by irradiation with UV photons (or another non-ergodic dissociation technique), allowing an opportunity to interpret the internal structure of proteins by measuring the collision cross sections of their native-like fragment ions.

[0159] We hence probed if considering collision cross-sections of native-like fragment ions could differentiate between two sets of ubiquitin conformations that are similarly compact but differ in their structure (Fig. 15a).

[0160] As outlined in Fig. 6, the cross-sections of these native-like protein fragments might sufficiently constrain the ubiquitin structure so that its complete and atomic-level structure could be revealed by computational analysis. Hence, we first tested if the collision crosssections of the native-like [£>36+2]4+, [y4o-2]4+, and [yss-2]5+fragment ions could allow differentiation between two MD-derived sets of ubiquitin models that are similarly compact but one set is composed of 836 predominantly native-like models whereas the other one comprises 2076 gas-phase ubiquitin models (Fig. 15A). For both sets of ubiquitin models, we calculated the cross sections for the intact ubiquitin chain and the respective [£>36+2]4+, [y4o-2]4+and [yss-2]5+fragment ions. The cross-section distributions for the native-like set centers at 1245 A2with a full-width-half-maximum (FWHM) of ~45 A2, whereas the gasphase annealed models center at a cross section of -1320 A2with a FWHM of -80 A2. Both sets of ubiquitin structures thus agree with the experimental cross-section of 1284 A2roughly equally (Fig. 15a).

[0161] Filtering both sets of structures against the collision cross section measured for intact ubiquitin selects models from both sets (Fig. 15b), results in an incorrect structure (Fig. 15d), a mean fraction of native contacts of Q-0.5, and a mean CaRMSD of -8 A. This underlines that knowledge of the collision cross section for intact ubiquitin is insufficient to rigorously differentiate between different protein structures that are similarly compact.

[0162] In contrast, by considering the experimental collision cross sections of the native-like fragment ions for structure assignment, we can exploit that the calculated collision cross sections of the fragment ions differ between the native-like and gas-phase annealed ubiquitin models. For structure assignment, we hence represented each model by a 4- dimensional vector of the relative deviation between the calculated and experimental cross sections (intact ubiquitin, [£>36+2]4+, [y4o-2]4+, [yss-2]5+) and used the Euclidean distance for identifying clusters. This approach correctly selects mainly models from the set of native ubiquitin conformations (Fig. 15b), leading to a mean fraction of native contacts of 0.85 and a mean CaRMSD of ~4.2 A (note that the RMSD reduces to 1 .5 A if the flexible C-terminal tail is not considered), and correctly assigns a strongly native structure assigned to the experimental data (Fig. 15c).

[0163] Fig. 16B outlines a general approach of how the fragment ion cross sections can be utilized to propose complete model structures for biopolymers.

[0164] Fig. 16C emphasizes that although the intact ubiquitin polypeptide models have similar cross-sections, the native-like and gas-phase models differ in the cross-sections of their fragment ions. To test if this difference sufficiently distinguishes the native-like and gasphase annealed models, we pooled all model structures, represented each model by the cross-sections of the intact ubiquitin polypeptide and the corresponding £>36, / 4o, and yss fragments, and performed density-based clustering of the models using an Euclidean distance metric. This produced a cluster of model structures that are overall consistent with the ubiquitin X-ray structure (median RMSD = 3 A, median fraction of native contacts Q = 0.9) that is separated from a cluster of mostly gas-phase models (median RMSD = 12 A, median Q = 0.1).

[0165] Next, we examined whether the measured fragment ion cross-sections would sufficiently constrain the ubiquitin structure so that the experimental cross-sections can be used to filter the correct ubiquitin structure from a pool of models that comprise mainly incorrect models. To test this, we generated a pool of 1363 native-like ubiquitin models and a set of 30989 decoy models using the available computational biology methods RoseTTAFold2 and Rosetta AbinitioRelax, computed the collision cross-sections for their intact polypeptide and the respective £>36, y4o, and yss fragments, and rank-ordered all models by their Euclidean distance d to the experimental cross-sections (Fig. 16C). We stress that these models contain a large number of partially correct folded models, which impedes the selection of correct models. Nevertheless, Fig. 16C shows that filtering all models within an error of + / - 1.5% selected 24 ubiquitin models at a precision of -54%, a median fraction of native contacts Q of ~0.9, and a median RMSD of 2.3 A. By contrast, filtering all models against the cross-section of only the intact polypeptide at an error of + / - 1.5% selected 1589 models at a precision of -27%, a median fraction of native contacts Q of -0.6 and a median RMSD of 6.7 A (Fig. 16C). Hence, our data underline that the fragment cross-sections sufficiently constrain the ubiquitin structure to filter out, on average, a largely native ubiquitin structure from a large pool of predominantly decoy models.

[0166] Encouraged by these results, we investigated if the measured cross-sections of the ubiquitin fragments could be used as features in machine-learning models that identify native-like ubiquitin models from the pooled set of the native and decoy models used above. To test this, we trained a Gaussian process on the set of 2927 native-like and gas-phase annealed MD-derived ubiquitin models (Fig. 16A) and validated the trained Gaussian process on the Rosetta-derived data set shown in Fig. 16C. The trained Gaussian process classified 227 models as ‘native’ at a precision of 82% and a recall of 14% (Fig. 16D). The median fraction of native contacts of these selected models is -0.99, the median RMSD is 0.7 A, and the structure averaged over the selected models (Fig. 16D) exhibits a fraction of native contacts of -1.0 and a root-mean-square distance of -0.8 A with respect to the x-ray structure. Taken together, our data show (i) that UV photodissociation in a tandem-trapped ion mobility spectrometer dissociates the protein ubiquitin into fragment ions with native-like structures and (ii) that the cross-sections of these native-like protein fragments sufficiently constrain the ubiquitin structure so that its atomic-level structure can be revealed by computational analysis.

[0167] Discussion

[0168] How generalizable are our results on ubiquitin presented. The critical aspects of our approach are (i) sequence coverage upon dissociation (e.g. UV photodissociation) in the sense that the fidelity to “resolve” a protein structure requires the generation of several native-like fragments, (ii) the ability of the produced fragment ions to maintain native-like structures, (iii) the availability of computational methods that propose putative protein structures that include the correct structure. Protein sequences in the UniProtKB / SwissProt database contain a median of 12 proline residues that could constitute cleavage sites for generating native-like protein fragments using our methodology, and 90% of these putative proline cleavage sites can be expected to result in fragments longer than 40 residues. Hence, because proline residues readily dissociate upon UV irradiation and produce intense [yn-2] ions, the vast majority of proteins should yield even more fragment ions than ubiquitin constrain their structures. Proline residues are predominantly found in turns and loops that connect polypeptide segments with secondary structure elements stabilized by hydrogen bonds. Hence, the majority of these putative protein fragments produced from cleavage proximal to proline residues should retain native-like structures in line with our above results on ubiquitin. Finally, deep neural networks such as AlphaFold and RoseTTAFold can be expected to generate a suitable set of putative model structures for protein systems, including for liganded or post-translationally modified protein complexes, which can then be filtered against the measured cross-sections as shown in Fig. 16. Consequently, the methodology described here can be expected to constitute a general approach to characterizing the structure of a protein system, including for liganded or post-translationally modified protein complexes. Given the high sensitivity and throughput of mass spectrometry-based methods, our approach is thus expected to enable protein structure characterization on a minute time scale and from complex and heterogeneous samples, including the presence of differentially modified proteins or multi-protein complexes.

[0169] These are important attributes because, arising from mechanisms such as alternative splicing of transcripts and post-translational modification of proteins, a typical mammalian cell is estimated to comprise over one million different proteoforms that can vary over time and of which only a minuscule number has been identified. The interaction between proteoforms, mediated at the molecular level through the (transient) formation of protein complexes, gives rise to protein-protein interaction networks that collectively regulate the function of a cell. This complexity takes on increased significance for structural prediction because of the sheer number of 300 common post-translational modifications coupled with cross-talk between multiple modifications on the same protein molecule and the lack of exhaustive structural data for the training of deep learning models. However, mass spectrometry methods are well-suited to handle the structural heterogeneity arising from proteoforms and assembly steady-states and enable systematic measurements of proteomes. Hence, the approach described here can be expected to be a significant step towards systematically discovering, on a proteome-wide scale and at the molecular level, how structural differences between proteoforms modulate protein interaction networks related to diseases.

[0170] Materials and Methods:

[0171] Sample preparation and introduction

[0172] Ubiquitin from bovine erythrocytes (> 98 %) and water (LC / MS grade) were obtained from Sigma-Aldrich (St. Louis, MO). Acetic acid (glacial) was obtained from Fisher Scientific (Pittsburgh, PA). High-concentration ESI tuning mix was obtained from Agilent (Santa Clara, CA). Following literature reports, we prepared an aqueous ubiquitin solution at a concentration of 10 pM with 1v% acetic acid for the measurements. ESI tuning mix was used as obtained for mass and ion mobility calibration. Tandem-trapped ion mobility spectrometer

[0173] All ion mobility experiments were performed on an orthogonal tandem-TIMS / MS instrument coupled with a 213 nm UV laser produced from the fifth harmonic of a Nd:YAG laser that was previously described. Briefly, the orthogonal tandem-TIMS / MS was constructed from modifying a commercially available timsTOF Pro instrument (Bruker Daltonics, Billerica, MA) by incorporating an additional trapped ion mobility spectrometry (TIMS) device between the electrospray capillary and the timsTOF Pro and a linear ion trap operating at 2-3 mbar between the two TIMS devices. Separation in TIMS is performed by linearly increasing the entrance potentials of the mobility analyzer region in a TIMS. For all measurements reported here, we increased the entrance potential in TIMS-1 over a time window of 50 ms and in TIMS-2 over a 100 ms time window. Ions elute from TIMS-1 according to their mobilities and can be selected (by timing the potentials between L1 and L2) and energetically activated either by means of energetic collisions with buffer gas neutrals (by timing the potentials between L3 and V1) and / or irradiation by UV photons. For UV photoactivation, the selected ions are stored in the linear ion trap for 50 ms. Mass analysis is conducted by the time-of-flight mass analyzer downstream of TIMS-2. Entrance and exit pressures of TIMS-1 and TIMS-2 were kept at approximately 4.7 mbar and 4.1 mbar (TIMS-1) and 2.2 mbar and 0.9 mbar (TIMS-2), respectively.

[0174] Collision-induced dissociation - collision-induced unfolding measurements of mobility-selected ubiquitin ions

[0175] The ubiquitin sample is injected into the electrospray ionization (ESI) source in positive mode via a gastight syringe at a flow rate of 180 .L / min. The generated ubiquitin ions enter the mobility analysis region of TIMS-1 and are separated based on their ion mobilities. As the mobility-separated ubiquitin ions elute from TIMS-1 according to their ion mobilities, the desired 8+ ubiquitin conformation is selected by timing the electric potentials at apertures L1 and L2, either by allowing their transmission by applying weakly accelerating electric fields or by blocking ions other than the desired conformation via repulsive electric fields. The mobility-selected ubiquitin conformation is then collisionally dissociated (CID) by applying a voltage difference of 180 V between apertures L3 and V1. Subsequently, the resulting CID fragment ions are transmitted through the linear ion trap and are transferred into the accumulation region in the TIMS-2 analyzer for 100 ms. The ions can be collisionally-unfolded by means of energetic ion-neutral collisions between the T2-T3 electrodes in the TIMS-2 analyzer ascribed elsewhere (collision-induced unfolding / CIU). To this end, we applied potentials from 5V (non-activating) to 140 V (strongly-activating) at intervals of 10 V. Finally, the ions are mobility-separated in TIMS-2 by raising the potential at the entrance of the mobility analysis region from -120 V to 5 V at a rate of 1.25 V ms-1. RF peak-to-peak voltages in TIMS-1 , multipole ion trap, and TIMS-2 were set to -290 V, -240 V, and -150 V.

[0176] UVPD-CIU measurements of mobility-selected ubiquitin ions

[0177] Ion generation, electrospray ionization, ion separation, and selection of the desired conformation were conducted. Subsequently, the selected ubiquitin ions were stored in the linear ion trap for 0 ms to 500 ms by timing electric potentials on electrodes V2 and V3. Ultraviolet photodissociation (UVPD) is performed on the stored ions by irradiation with -0 to - 500 laser pulses at a wavelength of 213 nm. Subsequently, the generated fragment ions were eluted from the ion trap and accumulated in the TIMS-2 analyzer for 100 ms. Electric potentials from 5V to 140 V at intervals of 10 V were applied between electrodes T2 and T3 in the TIMS-2 analyzer to collisionally induce the unfolding of the UV-produced fragment ions. Finally, the mobility of the ions is analyzed in TIMS-2.

[0178] Data Calibration

[0179] Mass and collision cross sections in orthogonal tandem TIMS / MS are determined via a calibration procedure. The mass and mobility calibrations were performed by using three calibration points from the ESI tuning mix using the reported reference mass and mobility values. By calibrating the system, measured masses and mobilities are assigned to corresponding reference masses and mobilities from the loaded reference list.

[0180] Differentiation between different, but equally compact, ubiquitin conformations: Structural ensembles

[0181] We used the two structural ensembles examined in Figure 16A to assess whether fragment ion cross sections enable differentiation between different, but equally compact, ubiquitin conformations.

[0182] (a) To represent native-like structures, we used the ensemble of 836 ubiquitin models that was computed by the SRA for charge state 8+ from an explicit-solvent molecular dynamics simulation and reported previously. Of these models, 564 are labeled ‘native’ structures as defined here by a fraction of native contacts greater than 0.8. The corresponding ion mobility spectrum shows a main feature centered at 1254 A2, corresponding to a mean fraction of native contacts of -0.9. Overall, the mean and median fraction of native contacts are -0.8 and -0.9, respectively.

[0183] (b) To represent non-native ubiquitin structures, we aggregated two previously used ensembles computed by the SRA, with 373 model structures derived from methanol / water explicit-solvent simulation of the ubiquitin A-state and 1703 model structures derived from gas-phase modeling of unfolded ubiquitin. The corresponding ion mobility spectrum shows a main feature centered at 1272 A2and corresponds to a mean fraction of native contacts of ~0.2.

[0184] Cluster analysis / Unsupervised learning

[0185] (1) For each model structure of the ensembles under S2.4.1 , we calculated the structures and cross sections for the £>36, y4o, and yss fragment ions as described in Section S2.2. above, and

[0186] (2) We pooled all model structures for which we successfully obtained cross-sections for the intact polypeptide chain (fl0) and the corresponding fragment ions (nb36, ny40, ny58).

[0187] (3) Each model was represented using the cross sections as features, i.e. as a tuple were calculated with respect to the experimental cross sections for intact ubiquitin (1283 A2) and the £>36 (790 A2), / 4o (83O A2), and yss (1010 A2) fragments fi =ncaic /

[0188] / A iexp

[0189] (5) The distributions were normalized to the intervals [0,... ,1],

[0190] (6) We performed locally scaled density-based clustering (LSDBC)on the models with a scaling of 0.25 (a) and considering 35 neighboring points ( ).

[0191] Supervised Learninq / Traininq of a binary classifier based on a Gaussian Process classifier:

[0192] Training set.

[0193] We used the pooled models used for cluster analysis to train the binary classifier. For classification, examples with a fraction of native contacts greater than 0.8 for the intact polypeptide were labeled ‘native’ and ‘non-native’ otherwise. This data set contained 2912 models with 564 models labeled ‘native.’

[0194] Testing set.

[0195] Generation of ubiquitin models.

[0196] The purpose of this testing set is to probe the reliability and generalizability of our approach by testing the trained Gaussian Process against an independent data set composed of

[0197] 1) Only a small set of native models (here -4%) among a large set of non-native structures.

[0198] 2) There is a significant number of partially correct models, e.g. with correctly folded secondary structure elements of the fragments but incorrect folding between these secondary structure elements.

[0199] 3) A significant number of partially folded models with similar cross-sections as the ‘native’ models.

[0200] For testing the trained classifier, we thus generated a set of 32352 ubiquitin models using the RoseTTAFoid2 and Rosetta Abini ti oRelax modules from the Rosetta suite of programs, of which 1363 exhibited a fraction of native contacts greater than 0.8 and were labeled ‘native.’

[0201] Abinitio relax. We executed Abinitio relax from the Rosetta 3.13 suite with the bovine ubiquitin sequence and the predicted secondary structure elements (PSIPRED) as input and the recommended settings of 10 cycles, re-weighting by radius of gyration (0.5), helix (0.5), and loops (0.5).

[0202] (a) Native and partially native structures. To generate closely native and partially native ubiquitin models, we successively executed Abinitio relax until a model was obtained with a fraction of native contacts Q > 0.8. This step was repeated roughly 250,000 times until a set of 789 models was accumulated.

[0203] (b) Non-native structures. To generate non-native ubiquitin models, we executed Abinitio relax for 29,802 times.

[0204] RoseTTAFoid2. As input, we used the ubiquitin amino acid sequence and ‘mmseqs_2’ as the multiple sequence alignment method. To obtain models with some structural variance, we set num_recycles to 0 and predicted 25 models for all possible oligomeric states (i.e. monomer through dodecamer).

[0205] Modeling of cross sections

[0206] (1) We removed steric clashes in the obtained models by performing an energy minimization and a subsequent 100 ps molecular dynamics simulation with GROMACS 4.5.7 using the OPLS / AA force field in conjunction with the Onufriev-Bashford-Case Generalized Born solvation model.

[0207] (2) Each model structure was subjected to molecular dynamics simulations and crosssection calculations using the procedure outlines under S2.1 and S2.2.

[0208] (3) Subsequently, relative cross-sections fij were calculated with respect to the experimental cross sections for intact ubiquitin (1283 A2) and the £>36 (790 A2), y4o (830 A2), and yss (1010 A2) fragments

[0209] Gaussian process classifier

[0210] We trained a binary classifier using a Gaussian Process. As kernel function, we chose the radial basis function with a characteristic length scale of 0.15 (note that the decision function appears independent of the length scale for the data sets studied here). The relative cross sections for the training and test data set were standardized, i.e., the mean was removed and the data scaled to unit variance before training and testing. LIST OF REFERENCE SIGNS

[0211] 1-4 quadrants of the TIMS tube

[0212] 6 ions

[0213] 7 gas

[0214] 8 capillary

[0215] 9 repeller plate

[0216] 10 entrance funnel

[0217] 11 TIMS tube

[0218] 11a accumulation unit of 11

[0219] 11b scan unit of 11

[0220] 13 exit funnel

[0221] 14 axial gas flow

[0222] 16 maximum gas flow velocity

[0223] 20 z location, beginning of rising edge

[0224] 23 z location, end of rising edge

[0225] 24 z location

[0226] 31 diaphragm location

[0227] 32 starting field point of gradient in 11 b

[0228] 33 end field point of gradient in 11b

[0229] 34 diaphragm location

[0230] 50 first field barrier

[0231] 51 second field barrier 52-59 ions with decreasing mobility

[0232] CID collision induced dissociation

[0233] CIU collision induced unfolding

[0234] ESI electrospray ionization

[0235] ED electron dissociation

[0236] IMS ion mobility separator otof orthogonal time of flight spectrometer

[0237] SID surface induced dissociation

[0238] TIMS trapped ion mobility spectroscopy

[0239] UVPD UV induced photo dissociation

Claims

CLAIMS1. Method for the determination of the conformation of a biopolymer, preferably by using hyphenating ion mobility mass spectroscopy, preferably downstream of a chromatographic or electrophoretic separation, wherein a) ionization, in particular electrospray ionization (ESI), preferably downstream of the chromatographic or electrophoretic separation, is used for generating native or essentially native conformation ions of said of at least one biopolymer in the gas phase; b) these ions, or precursors based on these ions, are selected, preferably based on at least one of mobility, charge state, or a combination thereof, c) these selected ions or precursors are fragmented using a non-ergodic or essentially non-ergodic dissociation to form fragments; d1) the resulting fragments are on the one hand analyzed, without further activation, for their distribution as a function of mass to charge, preferably in a survey spectrum, using a selection based on at least one of mobility, charge state, or a combination thereof; d2) the resulting fragments are on the other hand analyzed, in that they are either collision activated and then analyzed for their distribution as a function of mobility to charge and / or mass to charge in at least one spectrum as a function of said collision activation, or measured from further precursors having a different conformation, e) using differential information from d1) and d2) about the fragments to determine which of the fragments resulting from step c) are in a native or essentially native conformation, and using information from the fragments having native conformation for the determination of the conformation of said at least one biopolymer.

2. Method according to claim 1 , wherein in step c), the non-ergodic or essentially non-ergodic dissociation is selected from the group consisting of: ultraviolet photo dissociation (UVPD) or electron dissociation (ED), in particular electron capture dissociation.

3. Method according to any of the preceding claims, wherein in step c), the non- ergodic or essentially non-ergodic dissociation is selected as ultraviolet photo dissociation (UVPD) using a laser, preferably at a frequency in the range of 150-350 nm, preferably in the range of 200-250 nm, and / or preferably for a time span of at least 10 ms, preferably ofat least 20 ms or at least 40 ms, further preferably less than 1000 ms.

4. Method according to any of the preceding claims, wherein in at least one or both of steps b) and d), selection is carried out based on at least one of mobility, mass, charge state or a combination thereof, wherein preferably selection is carried out by introducing ions into an ion mobility separator (IMS), preferably a trapped ion mobility spectrometry (TIMS) separator.

5. Method according to claim 4, wherein in both of steps b) and d), selection is carried out by introducing ions into a trapped ion mobility spectrometry (TIMS) separator, wherein an ion trap is located between the two trapped ion mobility spectrometry (TIMS) separators, and wherein step c) is carried out while the respective ion is in the ion trap.

6. Method according to any of the preceding claims, wherein collision activation of step d2) is carried out via a multitude of activation experiments with variable collision energy to form collision induced unfolding curves of the fragment ions and / or is carried out while the respective ion is located in an ion trap downstream of step b) or, preferably, is carried out in the device used for selecting in step d2), wherein preferably said the device used for selecting in step d2) is a trapped ion mobility spectrometry (TIMS) separator.

7. Method according to claim 6, wherein the device used for selecting in step d1) is a trapped ion mobility spectrometry (TIMS) separator, and wherein at least in an accumulation section (11a) thereof collision induced unfolding is carried out by gas collisions, preferably by applying voltages of at least 5 V or at least 50 V or in the range of 100-200 V.

8. Method according to any of the preceding claims, wherein fragments with native or essentially native conformation are identified by the fact that the respective fragments are reducing or disappearing from the spectrum as a function of collision activation, and wherein for the determination of the conformation of the at least one biopolymer the collision cross-section measured of that fragment in d1) and / or d2) is used, preferably assisted by providing database information and / or molecular simulation information, in particular based on the primary structure of that fragment.

9. Method according to any of the preceding claims, wherein the biopolymer is selected from the group of proteins, peptides, nucleotides, polysaccharides, naturalrubbers, lignin, suberin, cutin, cutan, melanin, polyhydroxyalkanoates as well as mixtures and derivatives thereof, wherein preferably as a starting material for the biopolymer conformation determination also a mixture can be used, including proteomics samples such as plasma and urine samples.

10. Method according to any of the preceding claims, wherein at least one or in both of steps b) and d), selection is carried out by introducing ions into a trapped ion mobility spectrometry (TIMS) separator, wherein said trapped ion mobility spectrometry (TIMS) separator is a TIMS analyzer with parallel accumulation and separation, operating using a method comprising the steps:(a) accumulating ions entering in an RF ion trap;(b) transferring at least a subset of the accumulated ions into a trapping ion mobility separator, in which the transferred ions are radially confined by an RF field and pushed by a gas flow against a rising edge of an axial electric DC field barrier such that the transferred ions are spatially separated along the rising edge according to ion mobility;(c) successively releasing the transferred ions according to their ion mobility by decreasing the height of the electric DC field barrier while ions from the ion source are further accumulated in the RF ion trap; and(d) restoring the height of the electric DC field barrier which triggers a consecutive transfer of the accumulated ions from the RF ion trap into the trapping ion mobility separator.11 . Method according to any of the preceding claims, wherein at least one or in both of steps b) and d), selection is carried out by introducing ions into a trapped ion mobility spectrometry (TIMS) separator, wherein said trapped ion mobility spectrometry (TIMS) separator is a TIMS analyzer, and wherein separating ions in time according to mobility in the TIMS analyzer involves driving the ions transferred to the TIMS analyzer by a first gas flow against a first counter-acting electric DC field barrier such that the ions are trapped and spatially separated according to their mobilities at different positions along a ramp of the first electric DC field barrier at which a friction force of the first gas flow equals the counter-acting force of the first electric DC field barrier, and temporally separating ions according to mobility in the TIMS analyzer by adjusting a height of the first electric DC field barrier or the velocity of the first gas flow.

12. An apparatus for the determination of the conformation of at least onebiopolymer by using hyphenating mass spectroscopy with chromatographic or electrophoretic separation, in particular for carrying out the method according to any of the preceding claims, said apparatus preferably comprising: optional means for chromatographic or electrophoretic separation; electrospray ionization (ESI), preferably downstream of the chromatographic or electrophoretic separation, for generating native or essentially native conformation ions of said of at least one biopolymer in the gas phase; a first ion mobility separator (IMS) for receiving and sequentially releasing said native or essentially native conformation ions of said of at least one biopolymer ions from said IMS according to their ion mobility, an ion trap downstream of said ion mobility separator (IMS) and means for fragmenting using a non-ergodic or essentially non-ergodic dissociation to form fragments; a second ion mobility separator (IMS) for receiving and sequentially releasing said fragments from said IMS according to their ion mobility and allowing for collision activation, an apparatus for carrying out a mass spectroscopy measurement on said fragment ions.

13. An apparatus according to claim 11 , wherein the first and second ion mobility separator is a TIMS analyzer, preferably a TIMS analyzer with parallel accumulation and separation, in particular operating using a method comprising the steps:(a) accumulating ions in an RF ion trap;(b) transferring at least a subset of the accumulated ions into a trapping ion mobility separator, in which the transferred ions are radially confined by an RF field and pushed by a gas flow against a rising edge of an axial electric DC field barrier such that the transferred ions are spatially separated along the rising edge according to ion mobility;(c) successively releasing the transferred ions according to their ion mobility by decreasing the height of the electric DC field barrier while ions from the ion source are further accumulated in the RF ion trap; and(d) restoring the height of the electric DC field barrier which triggers a consecutive transfer of the accumulated ions from the RF ion trap into the trapping ion mobility separator.

14. An apparatus according to any of the preceding claims 12 or 13, wherein the main axis of the first ion mobility separator is perpendicular to the main axis of the second ion mobility separator, and wherein a UV laser is provided for irradiating the ion trap for non- ergodic or essentially non-ergodic dissociation.

15. An apparatus according to any of claim 12 -14, wherein the first ion mobility separator is a TIMS analyzer, in which the ions are radially confined by an RF field and pushed by a gas flow against a rising edge of an axial electric DC field barrier such that the transferred ions are spatially separated along the rising edge according to ion mobility, and wherein the height of the electric DC field barrier is chosen such as to retain only ions having an ion mobility higher than said upper threshold value (17), and wherein the second ion mobility separator is a TIMS analyzer with parallel accumulation and separation, in which the transferred ions are accumulated in an accumulation section (11a) from the first ion mobility separator in an RF ion trap, in which the ions are radially confined by an RF field and pushed by a gas flow against a rising edge of an axial electric DC field barrier such that the transferred ions are spatially separated along the rising edge according to ion mobility, and wherein the height of the electric DC field barrier is chosen such as to retain only ions having an ion mobility higher than said lower threshold value (18) followed by transferring at least a subset of the accumulated ions into a trapping ion mobility separator in a separation section (11b), radially confined by an RF field and pushed by a gas flow against a rising edge of an axial electric DC field barrier such that the transferred ions are spatially separated along the rising edge according to ion mobility, and wherein the height of the electric DC field barrier at the beginning of the release is preferably chosen such as to retain only ions having an ion mobility higher than said lower threshold value (18), wherein preferably between the accumulation section (11a) and the separation section (11 b) of the second ion mobility separator a second ion gate is provided preventing ions having an ion mobility higher than said lower threshold value (18) to enter the separation section (11 b).

Images