Mass Spectrometry and Ion Mobility Characterization of Bioactive

Dec 12, 2016 - ... source temperature of 80 °C, and desolvation temperature of 150 °C. ... Table 1. Measured and Calculated Collision Cross Sections...
0 downloads 0 Views 1MB Size
Subscriber access provided by University of Colorado Boulder

Article

Mass Spectrometry and Ion Mobility Characterization of Bioactive Peptide - Synthetic Polymer Conjugates Ahlam Alalwiat, Wen Tang, Selim Geri#lio#lu, Matthew L. Becker, and Chrys Wesdemiotis Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.6b03553 • Publication Date (Web): 12 Dec 2016 Downloaded from http://pubs.acs.org on December 17, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Analytical Chemistry Revised, 17 November 2016

Mass Spectrometry and Ion Mobility Characterization of Bioactive Peptide - Synthetic Polymer Conjugates

Ahlam Alalwiat,† Wen Tang,‡ Selim Gerişlioğlu,† Matthew L. Becker,‡ and Chrys Wesdemiotis†‡*



Department of Chemistry, The University of Akron, Akron, Ohio, 44325, United States



Department of Polymer Science, The University of Akron, Akron, Ohio, 44325, United States

_________________________ * Corresponding author. E-mail: [email protected]; phone: (330) 972-7699; fax: (330) 972-6085.

1 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ABSTRACT: The bioconjugate BMP2-(PEO-HA)2, composed of a dendron with two monodisperse poly(ethylene oxide) (PEO) branches terminated by a hydroxyapatite binding peptide (HA), and a focal point substituted with a bone growth stimulating peptide (BMP2), has been comprehensively characterized by mass spectrometry (MS) methods, encompassing matrixassisted laser desorption ionization (MALDI), electrospray ionization (ESI), tandem mass spectrometry (MS2), and ion mobility mass spectrometry (IM-MS). MS2 experiments using different ion activation techniques validated the sequences of the synthetic, bioactive peptides HA and BMP2, which contained highly basic amino acid residues either at the N-terminus (BMP2) or C-terminus (HA). Application of MALDI-MS, ESI-MS, and IM-MS to the polymerpeptide biomaterial confirmed its composition. Collision cross-section measurements and molecular modeling indicated that BMP2-(PEO-HA)2 exists in several folded and extended conformations, depending on the degree of protonation. Protonation of all basic sites of the hybrid material nearly doubles its conformational space and accessible surface area.

2 ACS Paragon Plus Environment

Page 2 of 27

Page 3 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

INTRODUCTION Hydroxyapatite (HA), a calcium phosphate bioactive material with the composition Ca10(PO4)6(OH)2, is widely used in bone tissue engineering because of its similarity with the mineral constituents of natural bone.1 Regeneration of bone tissue on HA or HA-coated scaffolds is stimulated by implanted stem cells which proliferate and differentiate into the desired tissue type that grows onto the scaffold and leads to bone healing.2 New tissue formation is significantly accelerated by adding growth factors, such as osteogenic growth peptide (OGP)3 or bone morphogenetic protein-2 (BMP-2),4 which promote the stem cell proliferation and differentiation processes.5 The most effective delivery of growth factors to the location they are needed is through vectors with high binding affinity to HA.6-9 This can be achieved by using hybrid materials functionalized with both an HA-binding substrate as well as the growth factor.7 Here, we present the complete characterization of the composition, sequence, and architecture of such a material using top-down, multidimensional mass spectrometry (MS) methodologies10,11 involving matrix assisted laser desorption ionization (MALDI)12,13 and electrospray ionization (ESI)14 interfaced with tandem mass spectrometry (MS2) and/or ion mobility mass spectrometry (IM-MS).15-17 The polymer-peptide based material investigated7 comprises two units of the peptide GGGSVSVGMKPSPRP which binds strongly to HA (abbreviated as HA peptide)6 as well as the peptide KIPKASSVPTELSAISTLYL which is the receptor binding site of BMP-2 (abbreviated as BMP2 peptide),18 linked through a dendritic poly(ethylene oxide) (PEO) construct (cf. Scheme 1). Such hybrid materials are difficult to prepare in crystalline or highly purified form for characterization by spectroscopic methods of analysis that probe average structures, such as NMR or X-ray diffraction.11 Because of MS’s dispersive nature, the multidimensional 3 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

approaches utilized here do not have such limitations, providing an alternative tool of structural identification/confirmation, as will be demonstrated for the peptides BMP2 and HA and their BMP2-(PEO-HA)2 conjugate. The approaches discussed would be equally applicable to biosimilars (biologic drugs that are similar but not identical with FDA-approved biologic drugs), which increasingly contain peptide(s) or protein(s) tethered to synthetic polymers. Recent studies have shown that gentle ionization methods (MALDI, ESI) can preserve solution conformation; in such cases, MS-based characterizations, which take place in the gas phase, also unveil valuable insight about biomolecular structure and folding in biological environments.19-21

EXPERIMENTAL SECTION Materials. The BMP2-(PEO-HA)2 hybrid material was prepared as outlined in Scheme 1.7 A brief description of the synthetic steps is given in the Supporting Information (SI) section.

Scheme 1. Synthetic route to hybrid material BMP2-(PEO-HA)2.7

4 ACS Paragon Plus Environment

Page 4 of 27

Page 5 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

The reagents and solvents needed for the syntheses and mass spectrometry analyses were purchased from Sigma-Aldrich (St. Louis, MO), VWR (Radnor, PA), or Fisher Scientific (Pittsburgh, PA); all were used in the condition received. MALDI-MS and MS2 experiments. MS and MS² experiments were performed in positive ion mode on a Bruker UltraFlex III MALDI tandem time-of-flight (ToF/ToF) mass spectrometer equipped with a Nd:YAG laser emitting at 355 nm (Bruker Daltonics, Billerica, MA). For the peptides BMP2 and HA, α-cyano-4-hydroxycinnamic acid (CHCA) was used as matrix. A solution of CHCA matrix was prepared in tetrahydrofuran (THF) at a concentration of 20 mg/mL and was employed to make the top and bottom layers of the samples deposited onto the MALDI target. Solutions of the peptides were prepared in H2O at a concentration of 20 mg/mL and were used to add the center layer of the samples (three-layer sandwich method).11,22 This sample preparation procedure led to the formation of abundant [BMP2 + H]+ and [HA + H]+ ions from both peptides. For the BMP2-(PEO-HA)2 hybrid material, sinapinic acid (SA) served as matrix. A solution of SA in aqueous acetonitrile (ACN/H2O, 70:30, v/v) was prepared at a concentration of 20 mg/mL; it was used to make the top and bottom layers of the sample subjected to MALDI. A BMP2-(PEO-HA)2 solution (10 mg/mL) was prepared in H2O. The two solutions were combined onto the MALDI target via the three-layer sandwich method. This sample preparation protocol led to the formation of singly and doubly protonated ions. The LIFT (laser-induced fragmentation) mode with no additional collision gas was employed for the acquisition of MALDI-MS2 spectra.23 Bruker’s flexAnalysis software was used for the analysis of MS and MS2 data.

5 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ESI-MS, IM-MS, and MS2 experiments. MS and MS² experiments were performed on a Waters Synapt HDMS quadrupole/time-of-flight (Q/ToF) mass spectrometer (Waters, Beverly, MA), equipped with an electrospray ionization (ESI) source and the traveling wave variant of IM-MS.24 The sample solution was introduced to the ESI source by direct infusion at a flow rate of 3 µL/min. The instrument was operated in positive ion mode with a capillary voltage of 3.15 kV, cone voltage of 35 V, sampling cone voltage of 3.2 V, source temperature of 80 °C, and desolvation temperature of 150 °C. The BMP-2 peptide was dissolved in H2O at a concentration of 0.05 mg/mL; 10% of methanol (MeOH) and 1% formic acid (both v/v) were added to this solution to improve sample ionization. The HA peptide and the BMP2-(PEO-HA)2 conjugate were dissolved in H2O at a concentration of 0.01 mg/mL; 10% of MeOH (v/v) was added to the solution of HA, and 30% of MeOH plus 0.5% formic acid (both v/v) were added to the solution of BMP2-(PEO-HA)2. The Synapt mass spectrometer contains three confined chambers in the interface region between the Q and ToF mass analyzers, called trap cell (closest to Q), IM cell, and transfer cell (closest to ToF).24 In IM-MS experiments, either all ions generated in the ESI source or ions within a specific m/z window selected by Q are separated by their mobilities in the IM cell before ToF mass analysis. Ion mobility separation was achieved by tuning the traveling wave height and traveling wave velocity applied to the IM cell at 8 V and 300 m/s for the peptides, or 8 V and 250 m/s for BMP2-(PEO-HA)2, respectively. The traveling wave height and wave velocity in the trap cell were set to 0.5 V and 300 m/s, respectively, and those in the transfer cell were set to 0.2 V and 248 m/s, respectively. The nitrogen gas (drift gas) flow rate in the IM cell was 22 L/h. MS2 experiments were performed in the transfer cell via collisionally activated dissociation

6 ACS Paragon Plus Environment

Page 6 of 27

Page 7 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

(CAD),25 using argon as collision gas. The MassLynx (version 4.1) and DriftScope (version 2.0) software was employed for data analysis. The peptides were also examined by ESI-MS2 experiments on a quadrupole ion trap (QIT) mass spectrometer (Bruker Daltonics, Billerica, MA), as explained in the SI section. This instrument permitted MS2 experiments via electron transfer dissociation (ETD)25 which are also presented in the SI section. Collision cross-section (CCS) determination. The conformation of the dendritic hybrid material BMP2-(PEO-HA)2 was investigated by acquiring the collision cross-sections of the different charge states observed upon ESI.11,15-17 For this, the drift time scale in the IM-MS experiments was calibrated using polyalanine standards.11,26 The data used to construct the calibration curve are summarized in Table S1. The curve obtained was validated by the determination of the known collision cross-sections of several charge states of ubiquitin and cytochrome C (cf. Table S2);27-29 the differences between published CCS and CCS determined in this study are 3-7%, with seven out of the eight pairs of CCSs lying within ≤5% (Table S2). The measured CCSs of the BMP2-(PEO-HA)2 bioconjugate with 4-9 proton charges are provided in Table S3 and compared with theoretical predictions in Table 1. Molecular modeling. Geometry optimization and energy minimization for different tautomers of BMP2-(PEO-HA)2 with 4-6 proton charges were performed by molecular mechanics/dynamics simulations with the Materials Studio program, version 7.0 (Accelrys Software, Inc.). The CCSs of the optimized structures were calculated using the MOBCAL algorithm (cf. Table S4).30 Details about the modeling procedure and CCS calculations are given in the SI section.

7 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

RESULTS AND DISCUSSION Characterization of BMP2 peptide. BMP2 is a hydrophilic peptide with twenty amino acid residues; it was prepared via solid state synthesis on a microwave peptide synthesizer.7 The sequence expected from this process is KIPKASSVPTELSAISTLYL (2117.20 Da monoisotopic mass).31 The MALDI-MS spectrum of BMP2 (Figure 1a) includes abundant peaks arising from [BMP2 + H]+ (m/z 2118.5), [BMP2 + Na]+ (m/z 2140.5), and [BMP2 + K]+ (m/z 2156.5) ions,

Figure 1. (a) MALDI-MS spectrum of BMP2 peptide; (b) MALDI-MS2 spectrum of protonated BMP2, viz. [BMP2 + H]+ (m/z 2118.5). The MS2 data confirm the sequence shown on the top, viz. KIPKASSVPTELSAISTLYL.

which agree well with the expected composition. The fragments present in the MALDI-MS2 spectrum of [BMP2 + H]+ (Figure 1b) confirm the sequence KIPKASSVPTELSAISTLYL. A contiguous and almost complete series of N-terminal bn ions is observed (from b3 to b19), resulting from cleavages of the C(=O)–NH amide bonds which coproduce C-terminal y-type ions 8 ACS Paragon Plus Environment

Page 8 of 27

Page 9 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

(cf. Scheme S1).32 The b-type ions dominate due to the presence of highly basic amino acids at and near the N-terminus, viz. K (residues 1 and 4) and P (residue 3).33 Consecutive dissociations to form internal fragments (designated by single letter code in Figure 1b) and an fragments via CO loss from bn also occur. Only two y-type ions are observed with significant intensity above noise level, viz. y18 and y12; the latter fragment is complementary to the much more abundant b8 ion. It is noticed in Figure 1b that the intensities of y18 and b8, which originate from cleavages of Xxx-Pro amide bonds, are noticeably high, reflecting enhanced amide bond cleavage at the Nterminal site of proline; such reactivity has been observed with many proline containing peptides34 and is attributed to the high proton affinity and secondary amine structure of this amino acid.35-37 BMP2 forms multiply charged ions under ESI conditions (Figure S1a) due to the inclusion of several highly basic amino acid units in its sequence. The most intense peaks correspond to [BMP2 + nH]n+ (n = 2-3), [BMP2 + 2H + X]3+ (X = Na or K), and [BMP2 + 3H + K]4+. Additionally, a few fragment ions are observed, most notably TELSA - 28, a2, and K. In IM-MS mode, the ESI-generated ions can be separated by their drift time in the IM chamber (IM dimension) and their m/z ratio in the ToF analyzer (MS dimension). Drift times depend on ion charge and collision cross-section (CCS); the latter parameter corresponds to the forward moving area of the ion in the IM chamber and is determined by the size (mass) and shape (conformation or architecture) of the ion.15 Generally, ions with higher charge states and smaller CCSs drift faster through the IM region than ions with lower charge states and larger CCSs.10,11,15-17 IM-MS analysis of BMP2 (Figure S1b) attests that such 2-D separation removes chemical noise and improves the dynamic range and sensitivity, allowing for the detection of singly charged peptide in addition to much more abundant doubly, triply, and quadruply charged 9 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

BMP2. The drift times measured for intact BMP2 in charge states 1+, 2+, 3+, and 4+, are 9.22, 5.26, 2.18, and 1.92 ms, respectively. The drift time distributions of these ions are monomodal, consisting of single, Gaussian peaks, as exemplified for mass-selected [BMP2 + 3H]3+ in Figure S1c; this result is consistent with one sequence and conformation for BMP2. Figure S2 shows the ESI-MS2 spectrum of mobility-separated [BMP2 + 3H]3+ (m/z 706.76), acquired via collisionally activated dissociation (CAD). Most fragment ions are singly or doubly charged N-terminal bn sequence ions because of the presence of highly basic amino acids at and near the N-terminus of BMP2 (vide supra); however, the bn series is less complete and the number of internal fragments higher than upon MALDI-MS2, compromising sequence information. The multiply charged ions formed by ESI also permit investigation of the BMP2 sequence by electron transfer dissociation (ETD)38 which mainly causes cleavages at the NH–Cα bonds, resulting in N-terminal cn and C-terminal zn• fragment ions, depending on which piece retains the charge(s), cf. Scheme S2.38,39 In the ETD spectra of [BMP2 + 2H]2+ and [BMP2 + 3H]3+ (Figure S3), cn ions dominate due to the high proportion of basic amino acids at the Nterminus of BMP2. The largely contiguous c-type series observed in both spectra and the complementary z-type fragments produced from [BMP2 + 2H]2+ cover the entire peptide sequence. Overall, these ETD data and the ESI-CAD and MALDI-MS2 results discussed fully corroborate the designed BMP2 peptide sequence. Characterization of HA peptide. HA is a hydrophilic peptide containing fifteen amino acid residues with the putative sequence GGGSVSVGMKPSPRP (1411.72 Da monoisotopic mass); this sequence has been found to exhibit high-specificity binding to crystalline hydroxyapatite.6 10 ACS Paragon Plus Environment

Page 10 of 27

Page 11 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Figure 2. (a) ESI mass spectrum of HA; (b) 2-D IM-MS plot and (c) drift time distribution (IMMS mobilogram) of mass-selected [HA + H]+ (m/z 1412.9). Sign ‡ denotes ions with 2+ charges.

The most abundant peaks in the ESI-MS spectrum of HA (Figure 2a) correspond to doubly and triply charged HA, having the compositions [HA + 2H]2+ and [HA + 3H]3+, respectively. The intensity of singly charged [HA + H]+ is low. The other ions in the spectrum are mostly fragments with different charges. Figure S4 shows the 2-D IM-MS plot obtained by separating all ions formed upon ESI by their mobilities before mass analysis. The IM dimension removes chemical noise and separates the intact singly, doubly, and triply charged peptide from fragments and aggregates. The drift times of intact HA with one, two, and three proton charges are 12.18, 3.16, and 1.53 ms, respectively. Under the same IM-MS conditions, singly, doubly, and triply protonated BMP2 have drift times of 9.22, 5.26, and 2.18 ms, respectively (vide 11 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

supra). The mass (size) of HA is 33% smaller than that of BMP2; this difference is reflected by the drift times of the doubly and triply protonated peptides, which are 30-40% smaller for HA than BMP2. In sharp contrast, [HA + H]+ shows a 32% larger drift time than [BMP2 + H]+, strongly suggesting an extended, helical conformation with a large collision cross-section for singly protonated HA (where the charge is located at the basic residues near the C-terminus), but a much more compact, globular conformation with a smaller CCS for singly protonated BMP2 (where the charge must be located at the basic N-terminus).40 It is evident from Figure S4 that the [HA + H]+ ion overlaps with additional components of the same m/z ratio. This is more clearly visible in the 2-D IM-MS plot acquired after mass selection of the ions transmitted within the m/z 1412-1413 window (Figure 2b). This plot and the corresponding mobilogram (Figure 2c) reveal four distinct drift time distributions peaking at 12.18, 7.04, 5.42, and 3.88 ms. The mass spectra extracted from the mobility separated regions (cf. Figure S4) have isotope spacings that identify the ion with a 12.18-ms drift time as singly charged HA peptide, viz. [HA + H]+; the ion drifting at 7.04 ms as doubly charged HA peptide dimer (2825.5 Da), viz. [(HA)2 + 2H]2+; and the ions drifting at 5.42 and 3.50 ms as triply and quadruply charged HA peptide trimer (4238.2 Da) and HA peptide tetramer (5650.9 Da), respectively, viz. [(HA)3 + 3H]3+ and [(HA)4 + 4H]4+. The dimer, trimer, and tetramer could be formed in solution or during the ESI process. The high tendency of HA to aggregate is attributed to the sequence of this peptide, which appears to promote strong interchain hydrogen bonding interactions that create the observed multimers.41,42 The sequence of HA was also examined by ESI-MS2 experiments on the doubly and triply protonated species, which are the most abundant molecular ions formed by ESI (cf. Figure 2a). Figure 3 shows the CAD spectrum of the mobility-separated [HA + 2H]2+ ion (m/z 706.9). 12 ACS Paragon Plus Environment

Page 12 of 27

Page 13 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

A contiguous series of singly charged C-terminal yn ions dominates (y3-y13), along with a few Nterminal and internal fragments which verify the sequence GGGSVSVGMKPSPRP for the HA peptide. On the other hand, the ETD spectrum acquired from [HA + 3H]3+ (m/z 471.6) (Figure S5) shows partially contiguous cn (n = 5-13) and zn• (n = 2-12) fragment series as well as a few yn fragments, all of which agree well with the sequence GGGSVSVGMKPSPRP.

Figure 3. ESI-IM-MS2 (CAD) spectrum of [HA + 2H]2+ (m/z 706.95). The mass-selected ions were separated by their ion mobilities and the component drifting at 3.16 ms (cf. Figure S6) was subjected to collisionally activated dissociation, followed by ToF mass analysis of the CAD products. The sequence coverage is indicated on the structure shown on top of the spectrum.

The MALDI-MS spectrum of HA (Figure S6a) contains abundant peaks arising from the protonated peptide, [HA + H]+ (m/z 1412.7), and an ion at m/z 1428.7 which is 16 Da heavier than [HA + H]+ and consistent with the composition [HA + O + H]+. Tandem mass spectrometry

13 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

helps to identify the sequences of both ions. The MALDI-MS2 spectrum of [HA + H]+ (Figure S6b) includes an almost complete yn series (n = 2-11), as well as a few bn fragments (b8, b14, and b14 + H2O), all of which are generated by amide bond cleavages (cf. Scheme S1). The Cterminal yn series dominates due to the presence of highly basic Arg and Pro residues at and near the C-terminus. The fragmentation pattern in the MALDI-MS2 spectrum conclusively validates that the HA peptide has the sequence GGGSVSVGMKPSPRP. The MALDI-MS2 spectrum of [HA + O + H]⁺ (m/z 1428.7) is shown in Figure S7. The dominant fragment at m/z 1365 is formed by loss of a 64-Da moiety, which most likely involves the elimination of CH3SOH from an oxidized methionine side chain (residue no. 7 from the Cterminus). All other fragments are reconciled by consecutive bond cleavages in [HA + O + H CH3SOH]+, as rationalized in the structure on top of the spectrum in Figure S7. Since no oxidized peptide was detected by ESI (vide supra), the oxidation must have happened during MALDI; the MALDI matrix used has indeed been found to cause (photo) oxidation of methionine residues.43 Characterization of BMP2-(PEO-HA)2 bioconjugate. The dendritic PEO utilized in the synthesis of the hybrid material had monodisperse branch lengths, leading to a single product with the expected composition C263H443N69O84S2 and a monoisotopic mass of 5976.195 Da. Figure 4a shows the MALDI-MS spectrum of BMP2-(PEO-HA)2, in which two ions stand out, viz. [BMP2-(PEO-HA)2 + H]+ at m/z 5977.192 and [BMP2-(PEO-HA)2 + 2H]2+ at m/z 2989.098; the corresponding calculated m/z values, 5977.203 and 2989.098, respectively, agree excellently with the measured data, providing strong evidence that the polymer-peptide conjugate has the correct elemental composition. Doubly charged ions are rare in MALDI at the molecular size of this hybrid material, unless special structural features permit it. The relatively 14 ACS Paragon Plus Environment

Page 14 of 27

Page 15 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

high intensity of 2+ ions is attributed to the inclusion of highly basic amino acid residues on each of the three peptide branches.

Figure 4. (a) MALDI-MS and (b) ESI-MS spectra of BMP2-(PEO-HA)2 hybrid material.

BMP2-(PEO-HA)2 forms a series of multiply charged ions with 4+ to 9+ proton charges under ESI conditions (Figure 4b); these are observed at m/z 1495.0475 (4+), 1196.2414 (5+), 997.0446 (6+), 854.7367 (7+), 748.0417 (8+), and 665.0270 (9+) and agree very well with the corresponding calculated m/z values of 1495.0567 (4+), 1196.2469 (5+), 997.0404 (6+), 854.7500 (7+), 748.0323 (8+), and 665.0295 (9+), respectively, further substantiating the composition of the hybrid material. The most intense peaks arise from charge states 6+, 7+, and 8+ when formic acid is added to the sample analyzed (see Experimental). Without adding formic acid, the most intense peaks shift to 5+, 6+, and 7+, pointing out that protonation takes place both in solution as well as during the ESI process.44,45 15 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Upon IM-MS analysis (Figure 5a), each charge state of intact BMP2-(PEO-HA)2 (from 4+ to 9+) is dispersed into a unique 2-D location, defined by a specific m/z ratio and drift time. Figure 5b includes the drift times of the multiply protonated BMP2-(PEO-HA)2 ions observed and the corresponding drift time distributions. The data in Figure 5b reveal the expected trend for charge states 9+ to 6+: as the charge decreases, the drift time increases due to decreasing drifting velocity; however, further decreases in the charge state, to 5+ and 4+, cause drift time decreases instead. This reversal diagnoses a conformational charge, brought upon by folding of BMP2-(PEO-HA)2 into a more compact structure, as charge repulsion forces are removed.46 The conformational changes occurring in BMP2-(PEO-HA)2 upon successive proton addition were examined more closely by converting the measured drift times of the different charge states to collision cross-sections. With the traveling wave variant of IM-MS utilized in our study, unknown CCS values are obtained through calibration of the drift time scale with

Figure 5. (a) 2-D IM-MS plot of BMP2-(PEO-HA)2 hybrid material; (b) drift time distributions (IM-MS mobilograms) of the intact BMP2-(PEO-HA)2 charge states (charges supplied by H+).

16 ACS Paragon Plus Environment

Page 16 of 27

Page 17 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

standards of known CCS.26 Figure S8 depicts the calibration plot built from singly and doubly protonated polyalanine oligomers and Table S1 lists the CCS values of these ions and explains the procedure for setting up the calibration curve.11 The reported CCSs of several charge states of ubiquitin and cytochrome C could be reproduced within ~5% with the curve in Figure S8 (cf. Table S2), validating its suitability for application to BMP2-(PEO-HA)2. Table S3 details the conversion of the drift times of protonated BMP2-(PEO-HA)2 ions to collision cross-sections; the latter values are also included in Table 1. Increasing the charge from 4+ to 9+ approximately doubles the CCS from 741 to 1463 Å2 (+ 97%). Most of the increase occurs upon adding a proton to the 4+ charge state (741 → 1070 Ų or +44%). A further proton addition (5+ → 6+) causes a smaller but still substantial CCS increase (1070 → 1328 Å2 or +24%). More than 6 protons exert only subtle charges in the BMP2-(PEO-HA)2 conformation (+2-5%). These trends strongly suggest that the three peptide chains of the hybrid material are tightly folded through a network of inter- and intrachain hydrogen bonds when four proton charges are added. Significant unfolding occurs when a 5th proton is added and more unfolding is caused when the 6th proton is added due to charge repulsion. Further proton additions cause very small CCS changes (~10% from 6+ to 9+), indicating that most hydrogen bonding between remote locations has been disrupted with 6 proton charges. Additional information on the structure of the BMP2-(PEO-HA)2 bioconjugate in charge states 4+, 5+ and +6 was sought by molecular mechanics/dynamics modeling. Proton charges were added to arginine and lysine residues which have the highest proton affinities among the common amino acids found in peptides and proteins (1051 and 996 kJ/mol, respectively),33 as well as to the triazole functionalities. The proton affinities of 1,4-substituted triazoles,47 like those present in BMP2-(PEO-HA)2, are higher than those of proline (921 kJ/mol), tyrosine (926 17 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 27

kJ/mol), or methionine (935 kJ/mol),33 which are the other relatively basic amino acids in BMP2(PEO-HA)2; hence, the triazoles are also likely protonation sites. Table 1. Measured and calculated collision cross-sections of BMP2-(PEO-HA)2 ions in different charge states (H+ charges) CCS (Å2) z

4

5

6

m/z

1495.248

1196.241

997.045

Proton charge location d

td (ms)

exp.

calcd.

a

b

c

3.08

741

790

Triazoles (N, N); HA1 (R); HA2 (R)

905

HA1 (K, R); HA2 (K, R)

923

BMP2 (K, K); HA1 (R); HA2 (R)

933

BMP2 (K); HA1 (R); HA2 (K, R)

887

BMP2 (K, K); HA1 (K, R); HA2 (R) e

1058

BMP2 (K, K); HA1 (K, R); HA2 (R) f

920

BMP2 (K); HA1 (K, R); HA2 (K, R)

868

BMP2 (K, K); Triazole (N); HA1 (K, R); HA2 (R) e

1163

BMP2 (K, K); Triazole (N); HA1 (K, R); HA2 (R) f

785

BMP2 (K, K); Triazole (N, N); HA1 (R); HA2 (R)

908

BMP2 (K, K); HA1 (K, R); HA2 (K, R)

4.14

1070

4.43

1328

7

854.737

3.43

1369

8

748.042

2.85

1431

9

665.027

2.34

1463

a

Measured drift time.

c

Calculated CCS using the projection approximation method for the structure shown on the adjacent

column.

d

b

Experimental CCS based on the calibration plot in Figure S8 (see also Table S3).

The protonated amino acid residue(s) in BMP2 peptide and the two HA peptides (HA1 and

HA2) are specified in parenthesis; the notations triazole (N) and triazole (N, N) indicate that one or both triazole rings in the PEO linker chains of BMP2-(PEO-HA)2 are protonated at an N atom. e

e,f

These

f

tautomers can fold into compact ( ) or extended ( ) conformations with distinct CCS (see text).

BMP2 peptide contains two lysine residues, each HA peptide contains one arginine and one lysine residue, and each PEO dendron branch contains one triazole moiety. The structures of

18 ACS Paragon Plus Environment

Page 19 of 27

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

several BMP2-(PEO-HA)2 conformers with 4-6 protons on these sites (described in Table 1) were optimized and the collision cross-sections of the resulting conformations were calculated by the projection approximation (PA), trajectory (TJ), and exact hard-sphere scattering (EHSS) methods of the MOBCAL program (cf. Table S4). The BMP2-(PEO-HA)2 hybrid material with 4-6 proton charges has 865-867 atoms. At this size, the TJ method is believed to provide the most reliable CCS predictions.30,48 On the other hand, the PA method usually underestimates CCSs of larger ions, especially if they have nonspherical shapes,48,49 while the EHSS method tends to overestimate CCSs of systems with