Protein StructureâFunction Correlation in Living Human Red Blood

Subscriber access provided by UNIV OF CAMBRIDGE

Article

Protein structure function correlation in living human red blood cells probed by isotope exchange based mass spectrometry Sreekala Narayanan, Gopa Mitra, Monita Muralidharan, Boby Mathew, and Amit Kumar Mandal Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.5b03217 • Publication Date (Web): 04 Nov 2015 Downloaded from http://pubs.acs.org on November 9, 2015

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 18

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

Protein structure function correlation in living human red blood cells probed by isotope exchange based mass spectrometry Sreekala Narayanan#, Gopa Mitra#, Monita Muralidharan, Boby Mathew, Amit K Mandal* Clinical Proteomics Unit, Division of Molecular Medicine, St. John’s Research Institute, St. John’s National Academy of Health Sciences, 100 ft Road, Koramangala, Bangalore -560034, India #

These authors contributed equally to this work

*Corresponding Author Amit Kumar Mandal Professor Clinical Proteomics Unit, Division of Molecular Medicine St. John’s Research Institute St. John’s National Academy of Health Sciences 100 ft Road, Koramangala Bangalore -560034, India Phone: +91-80-49467107 Fax: +91-80-25501088 E-mail: [email protected]

ACS Paragon Plus Environment


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: To gain insight into the underlying mechanisms of various biological events, it is important to study the structure function correlation of proteins within cells. The structural probes used in spectroscopic tools to investigate protein conformation are similar across all proteins. Therefore structural studies are restricted to purified proteins in vitro and these findings are extrapolated in cell to correlate their functions in vivo. However, due to cellular complexity, in vivo and in vitro environments are radically different. Here, we show a novel way to monitor the structural transition of human hemoglobin upon oxygen binding in living RBCs, using hydrogen/deuterium exchange based mass spectrometry (H/DX-MS). Exploiting permeability of D2O across cell membrane, the isotope exchange of polypeptide backbone amide hydrogens of hemoglobin was carried out inside RBCs and monitored using MALDIMS. To explore the conformational transition associated with oxygenation of hemoglobin in vivo, the isotope exchange kinetics was simplified using the method of initial rates. RBC might be considered as an in vivo system of pure hemoglobin. Thus, as a proof-of-concept, the observed results were correlated with structural transition of hemoglobin associated with its function established in vitro. This is the first report on structural changes of a protein upon ligand binding in its endogenous environment. The proposed method might be applicable to proteins in native state, irrespective of its location, concentration and size. The present in-cell approach opens a new avenue to unravel a plethora of biological processes like ligand binding, folding and post-translational modification of proteins in living cells.


Page 2 of 18

Page 3 of 18

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


Proteins inside a cell function in a highly crowded milieu of macromolecules. Proteinaceous as well as non-proteinaceous macromolecules such as actin filaments, ribosomes, polyethylene glycol (PEG), osmolytes etc. can notably impact the equilibria and kinetics of biochemical processes such as protein folding, binding, and oligomerization inside the cell.1 Various spectroscopic techniques, e.g. fluorescence, circular dichroism, infrared and nuclear magnetic resonance spectroscopy used to visualize protein structure are mostly intrinsic chromophore specific that are similar across all proteins present in a cell. X-ray crystallography that provides high resolution structural information requires proteins to crystallize. Therefore it is impossible to study the structure of individual protein in the surroundings of other macromolecules. The classical approach of protein structural elucidation using spectroscopic tools is thus limited to purified molecules in vitro. Unlike other techniques, mass spectrometry has the advantage of molecular specificity, which allows to delineate stability of proteins even in an unpurified milieu in vitro. One such technique is commonly known as SUPREX (Stability of Unpurified Proteins from the Rates of H/D Exchange), which monitors isotope exchange kinetics of polar hydrogens in proteins using mass spectrometry.2,3 The in vitro structural information obtained by spectroscopic tools is subsequently extrapolated in vivo to correlate its function. However due to cellular complexity, in vitro observation does not always reflect the true structure function paradigm in cell. Therefore, it is necessary to study the structure of proteins at residue level in their endogenous environment. Unfortunately there is no analytical technique available to monitor protein structure inside a cell. In the present study, exploiting permeability of D2O across cell membrane, the isotope exchange of polypeptide backbone amide hydrogens of hemoglobin was executed inside living red blood cells (RBCs). Subsequently, the structural transition of hemoglobin upon oxygenation in cell was monitored using hydrogen/deuterium exchange based mass spectrometry (H/DX-MS).



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 4 of 18

EXPERIMENTAL SECTION Venous blood, anticoagulated with EDTA, was collected from healthy volunteers with their prior consent. After centrifugation at 3000 rpm for 10 min at 25°C, plasma was removed. The pelleted RBCs were washed thrice with 0.9% aqueous NaCl. Preparation of deoxy-Hb and oxy-Hb. In general, a red blood cell contains a mixed population of deoxy and oxy hemoglobin, with the major proportion being the hemoglobin in oxy state. To study the protein individually in its deoxy as well as oxy state, it is important to convert all the hemoglobin molecules to the same state. Deoxy-Hb was prepared by suspending 267 µL of RBC in 4 mL potassium phosphate buffered D2O (pD 7.4) containing 25 mM sodium hydrosulfite (oxygen scavenger) and N2 gas was bubbled to maintain the deoxy conditions throughout the experiment. For oxy-Hb, 500 µL of RBC was suspended in 4 mL potassium phosphate buffer and O2 gas was bubbled for 1 h prior to isotope exchange. This solution was centrifuged at 12,000 rpm for 1.5 min. 267 µL of the pellet was added to 4 mL potassium phosphate buffered D2O with O2 gas bubbling throughout the experiment. Osmolarity was maintained at 300 milliosmolar in both the sets to prevent lysis of the cells. H/DX Experiment. The hydrogen/deuterium exchange of deoxy and oxy hemoglobin inside RBC was initiated by incubating the cells from the respective reaction sets in buffered D2O. To monitor H/DX kinetics, 90 µL aliquot of RBC suspension was taken from the isotope exchange reaction set at different time intervals and centrifuged at 12,000 rpm for 1.5 min. RBC lysis and H/DX quenching were achieved by adding 1 µL of pellet to 0.1% aqueous TFA solution (pH 2.5, 0°C). To reduce the back exchange of incorporated deuterium atoms with protons from solvent, the proteolytic digestion was performed in situ by immediate addition of 50 µL of 4 µM pepsin solution. The enzyme:substrate ratio was maintained at 1:50 mol/mol and digestion was performed for 5 min. Sample was centrifuged at 12,000 rpm for 1.5 min and the supernatant containing the exchanged globin peptides was isolated. 5 mg/mL of the matrix, αcyano-4-hydroxycinnamic acid was prepared in 2:8:1 acetonitrile/ethanol/0.1% aqueous TFA. Equal volumes of the supernatant and matrix solution were mixed. 1 µL of this mixture was spotted on MALDI plate and dried immediately using a vacuum desiccator. These conditions were followed to ensure minimal back exchange. Mass Spectrometry. MS analyses were performed on Waters Synapt HDMS equipped with Matrix Assisted Laser Desorption Ionisation (MALDI) source in the positive ion V mode, using 200 Hz solid state laser (λ = 355 nm). An average of 20 scans was combined for each spectrum. Mass Spectrometer was calibrated using PEG mix. Data analysis. The obtained mass spectra were baseline-corrected and the isotope averaged centroid mass of each molecular ion (Mt) at a given time ‘t’ was measured using the software HX Express version Beta (http://www.hxms.com/HXExpress).4 The number of deuterium incorporated into a peptide, D(t) at time ‘t’ was calculated as follows:5 D(t) = (Mt – Mo) × N (M∞ – Mo)

(1)

where Mo and M∞ are the observed centroid masses for an undeuterated and a fully deuterated molecule respectively, and N is the total number of backbone amide hydrogens in the peptide. To calculate the mass of the undeuterated peptide (Mo), the RBC lysis and digestion of


Page 5 of 18

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


hemoglobin was performed under quench conditions containing 200 picomoles of pepsin, followed by subsequent analysis. We have observed that the saturation of deuteration of Hb was achieved at 6 h of exchange reaction. To calculate the mass of the fully deuterated peptide (M∞), the isotope exchange reaction was continued for 6 h followed by lysis of the RBC, pepsin digestion of hemoglobin and analysis of the mass spectra. In large excess of D2O, the exchange of polar hydrogens in protein follows pseudo firstorder kinetics. Ideally, each amide hydrogen contributes distinctly to D(t) but in practice all backbone amide hydrogens (N) of a peptide can be grouped into three categories, fast (PA), intermediate (PB) and slow (PC) exchanging with respective rate constants k1, k2, k3. D(t) can be expressed as D(t) = N – PAe-k1t – PBe-k2t – PCe-k3t

(2)

In our analyses, the isotope exchange rates were quantitatively assessed by fitting the data obtained from equation 1 into equation 2. The exchange rate constants and the populations of amide hydrogens in different groups were obtained by minimizing the sum of the squared residuals (SSR) through multiple iterations using Microsoft Solver. The combination of kinetic parameters with lowest SSR provides the most reliable best fit values that can be calculated from a given dataset. The SSR for each data set was calculated using the following equation: SSR = Σi [(yiobs – yicalc)/ yiobs]2

(3)

yiobs and yicalc were calculated from equation 1 and equation 2, respectively.6 The kinetic parameters PA, PB, PC, k1, k2, and k3 for deoxy and oxy hemoglobin in vivo were obtained from the best fit curve of D(t) vs. ‘t’ plot. At a given temperature and pH, these parameters are driven by the intermolecular hydrogen bonding and solvent accessibility of backbone amide protons in the protein molecule. Though principally it is possible to translate the differences in the rate constants and populations between various groups to the conformational changes associated with oxygenation of hemoglobin, in practice it is difficult to deal with 12 such parameters simultaneously. To overcome such complexity in the analysis, we adopted a simplified approach. Using the method of initial rates, the rate of exchange for different groups of amide hydrogens was calculated from the product of rate constants and respective populations. Subsequently, the difference in exchange rates within a group between oxy-Hb and deoxy-Hb was calculated. Summation of differential rates of three groups indicated structural changes in the respective region of tetrameric hemoglobin on oxygenation. A positive sign implied increased flexibility whereas a negative sign implied increased rigidity. Sequencing of Peptic Peptides. Sequence assignment of different peptic peptides obtained from the globin chains was carried out using tandem mass spectrometry. MS/MS spectrum of each peptide, obtained from the undeuterated reaction set, was recorded by selecting the respective precursor ions and fragmenting them by increasing trap collision energy. The generated product ions were manually assigned to obtain sequence information of the precursor ion. The sequence coverage obtained from the identified peptides was 55.3% for α globin chain and 70% for β globin chain.



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 Human hemoglobin is a tetramer (α2β2) consisting of α and β globin chains in duplicate and its oxygenation is regulated by cooperative ligand binding and allosteric factors like pH and 2,3-diphosphoglycerate (DPG). Extensive in vitro research with purified hemoglobin established that the cooperativity and allosteric regulation operates through structural transition.7 Red blood cell can be considered as an in vivo system of pure native hemoglobin, since its hemoglobin content is 95% of the total protein. In the present study, isotope exchange of hemoglobin was executed inside cell by incubating fresh RBCs in buffered D2O maintaining the osmolarity of the cells at 300 milliosmolar. H/DX kinetics of oxy-Hb and deoxy-Hb was investigated after quenching the exchange reaction by reducing the temperature and pH at different time intervals followed by lysis of RBCs from the reaction set. The obtained hemolysate was proteolytically digested using pepsin under quenched conditions and the resultant peptides were analyzed by MALDIMS. A schematic representation of the above in-cell process is shown in Figure 1. The obtained kinetic data were analyzed using the method of initial rates to assess the conformational transition. Table 1 shows the predicted structural changes of several peptides across different regions of human hemoglobin on oxygenation in vivo. Deuterium incorporation kinetics of the proteolytic peptides reflected conformational changes associated with oxygenation of hemoglobin at different regions of the tetramer. Subsequently, the observed structural transition inside RBC was correlated with the reported structure function correlation of hemoglobin established in vitro. Among the fourteen peptides that were analyzed, six were found to have an increased flexibility on oxygenation (Table 1). The following six peptides with m/z 1869.1 (β130-146), 1494.9 (β1-14), 1585.9 (α34-46), 1921.0 (β86-102), 1308.7 (β32-41) and 1161.7 (β32-40) appeared in the increasing order of their flexibility. Eight peptic peptides with m/z 2051.1 (β111-129), 3428.0 (α110-141), 931.6 (β103-110), 1799.0 (β15-31), 2910.6 (α1-29), 1000.6 (β22-31), 3326.8 (α1-32) and 3473.9 (α1-33) were found to have an ascending order of increase in rigidity on deoxy to oxy transition (Table 1). Figure 2A and 2B show the representative deuterium incorporation kinetics of the peptide 1921.0 m/z for deoxy and oxy hemoglobin respectively. Figure 2C represents the best fit curve of exchange kinetics for the peptide in deoxy and oxy states, as obtained by the method described in the experimental section (equation 1 and equation 2). The envelopes for isotope exchange kinetics for all the other peptides have been provided in the supplementary information (S1-13). Figure 3 represents the best fit curves of exchange kinetics for these peptides in their deoxy and oxy states.


Page 6 of 18

Page 7 of 18

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


DISCUSSION Understanding biological events at its molecular level requires both structural analysis and functional correlation of proteins to be studied inside a cell. Exploiting permeability of D2O across RBC membrane, we have studied the structural transition of hemoglobin upon oxygenation inside living RBCs using H/DX-MS. Due to the presence of large excess of hemoglobin in RBC, it can be approximated that the structure function correlation of the molecule established in vitro is nearly identical to that in vivo. In the present study, we have correlated the obtained in vivo results with the reported structure function correlation of purified hemoglobin established in vitro. Oxygenation of hemoglobin reduces the spin state and thereby decreases the radius of 2+ Fe of heme unit. Subsequently Fe2+ is pulled into the plane of porphyrin ring. Fe2+ is covalently attached to globin chains with the proximal histidine residue of helix-F.8 Movement of Fe2+ causes displacement of F and H helices with a resultant reduction in the pocket between them, expelling penultimate tyrosine residue from the pocket. In our in vivo experiments, the proximal histidine was observed to be a part of the peptic fragment with m/z 1921.0. The increased flexibility of this region on oxygenation can be attributed to the several changes that occur during deoxy to oxy transition. In vitro studies showed that the expulsion of βTyr145 causes rupture of the hydrogen bond between βVal98-βTyr145 and the salt-bridge between βAsp94-βHis146.8,9 Additionally, hydrogen bonds between βAsp99-αTyr42 and βAsp99αAsn97 at subunit interface are weakened in oxy-Hb.9,10 The disappearance of above interactions and the increased sulfhydryl reactivity of βCys93 indicate an overall increase in the flexibility of the surrounding residues compared to the rigidity contributed by the hydrogen bond between βAsn102-αAsp94 in oxy-Hb.11,12 Therefore the observed increased flexibility of the concerned region in hemoglobin upon oxygenation in vivo corroborates well with the reported literature. 2,3-DPG is an allosteric regulator bound between two β subunits in deoxy-Hb through ionic interactions with βVal1, βHis2, βLys82 and βHis143 residues of both β chains.13 Oxygenation causes reduction in size of the DPG binding pocket followed by its dissociation, which makes all interacting residues free.13 Oxygen transport is also regulated by the Bohr effect. The disappearance of salt-bridges between βHis146-βAsp94 and βHis146-αLys40 in oxy-Hb result in decreased pK of the imidazolium hydrogen, followed by its dissociation at blood pH, known as Bohr effect.14 The release of CO2 from cellular metabolism decreases pH, which subsequently shifts the equilibrium from oxy to deoxy state. In oxy-Hb, βTyr145 dissociates from βVal98 and αThr41.8,10 βHis143, βTyr145 and βHis146 are part of the peptide with m/z 1869.1. Thus βHis143 and βTyr145 contribute towards increased flexibility whereas the salt-bridge between βAsn139-βArg104 contributes towards rigidity.15 Although βHis146 dissociates from βAsp94 and αLys40, it further gets associated with βLys144 in oxy-Hb.15 Therefore as a net effect, only a small increase in flexibility is expected on oxygenation among residues of this peptide, as observed in vivo. Two other DPG binding residues, βVal1 and βHis2, belonging to the peptide 1494.9 m/z (β1-14), contributes towards increased flexibility in oxy-Hb, whereas the salt-bridge between βGlu7-βLys132 imparts rigidity in this region.15 As discussed, the H/DX data analysis for this peptide showed an increased flexibility. Thus, the reported conformational changes support the H/DX data. In oxy-Hb, the movement of a functionally important residue αHis122, away from αAsp126 towards βArg30 results in decreased pK of the imidazolium hydrogen followed by the release of a Bohr proton.14 Also, the expulsion of the penultimate tyrosine of α globin, αTyr140, results in dissociation of αTyr140-αVal93, αArg141-αVal1 and αArg141-αAsp126



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

linkages.8,14 Additionally, the disappearance of other interactions of αArg141 with αAsp6, αLys127 and βVal34 in oxy-Hb imparts increased flexibility to the peptide with m/z 3428.0,10,14 whereas the salt-bridge between αHis112-αGlu27 and hydrogen bond between αHis122-βTyr35 contributes to rigidity.15,16 Overall, an increased rigidity was observed for this peptide from the H/DX data. Although the residual interactions indicate an increased flexibility, additional structural information is required to explain the observed data for the large peptide 3428.0. The disappearance of the salt-bridge αVal1-αArg141 in oxy-Hb results in decreased pK of N-terminal αNH2 and the subsequent release of a Bohr proton.14 The bond between αAsp6αLys127 breaks on oxygenation.14 αVal1 and αAsp6 are part of the three overlapping large peptides 2910.6 m/z, 3326.8 m/z and 3473.9 m/z where the H/DX data showed different extent of rigidity on oxygenation, as discussed. In oxy-Hb, other constituent residues of the three aforementioned peptides participate in the following salt-bridges: αLys7–αAsp74, αGlu27– αHis112 and αGlu30–αHis50.15 However due to lack of sufficient literature information, it is difficult to explain the varying degree of rigidity observed among these peptides. The α chain peptide with m/z 1585.9 is an important peptide as it is a part of the subunit interface of hemoglobin. Several residues of this peptide are involved in vital inter-subunit contacts which play a major role in the cooperative nature of hemoglobin. The perturbation of following inter-subunit interactions: αLys40-βHis146, αThr41-βTyr145, αTyr42-βAsp99 on oxygenation may have resulted in the increased flexibility to this peptide observed in our in vivo experiments. On oxygenation, three inter subunit contacts between βArg40-αLeu91, βArg40-αArg92 and βTrp37-αAsp94 are perturbed, whereas H-bonds between βArg40-αThr41 and βTyr35αHis122 are formed.10,16 Few of the residues involved in these interactions are part of the overlapping peptides 1161.7 m/z and 1308.7 m/z. Based on the reports published, an overall increase in flexibility can be expected for these peptides, which has been reflected in the observed results as well. The observed in vivo results for the β chain peptide with m/z 931.6 indicated an increase in rigidity for this peptide on oxygenation. This increased rigidity might have been contributed by the formation of salt-bridges between βArg104-βAsn139 and βAsn108-αHis103 in oxy-Hb.15,16 βArg104 and βAsn108 are part of this peptide. Here the observed results corroborated well with the previously published literature. According to the observed H/DX data, the peptide 2051.1 m/z exhibited a small increase in rigidity upon oxygen binding. Literature reports indicate the disappearance of the H-bond between βGln127-αArg3117 and formation of salt-bridge βGlu121-βLys17 on oxygenation.16 Based on the above reports, the expected conformational change for this peptide is negligible. Thus additional information regarding the structural changes in this region on oxygenation is required to explain the observed data. Increased rigidity that was observed for the two overlapping peptides, 1000.6 m/z and 1799.0 m/z on oxygenation can be attributed to certain structural perturbations in the vicinity of this region. In addition to the salt-bridge between βGlu121-βLys17, a decreased distance between βArg30-αHis122 in oxy-Hb supports the observed rigidity.14,15 However additional structural information is required to support the significant increase in the observed rigidity. The obtained in vivo structural information of hemoglobin correlated well with the established in vitro results. The present in-cell H/DX method is a novel approach to study the conformational dynamics of hemoglobin upon oxygenation inside RBC. To our knowledge, this is the first report on structural transition of a protein associated with ligand binding, monitored in its endogenous environment. In previously reported in-cell


Page 8 of 18

Page 9 of 18

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


studies using mass spectrometry, NMR, fluorescence tagging and X-ray crystallography, the experimental proteins were either overexpressed in different host cells or modified genetically/chemically.18-21 In a MS-based study, the thermodynamic stability of the Nterminal λ6-85 repressor protein was determined in presence of increasing concentrations of urea upto 3M.18 Here, the protein was overexpressed within E. coli cells and the subsequent analysis was done using hydrogen/deuterium exchange followed by MALDI mass spectrometry. The structure of the putative heavy-metal binding protein TTHA1718 was calculated under in vivo conditions using in-cell NMR spectroscopy.19 Here the protein from Thermus thermophilus was overexpressed in E.coli cells and the backbone and side-chain resonances were assigned to calculate the 3D structure. The stability and aggregation of the mammalian cellular retinoic acid-binding protein I was monitored in vivo by real-time fluorescent labeling.20 However, the protein was suitably mutated and overexpressed in E.coli cells. Crystals of Trypanosoma brucei enzyme Cathepsin B were grown in vivo and its structure and activity was determined using X-ray crystallography.21 Here, the polyhedron gene was replaced by site-specific transposition of the gene encoding the protein of interest in the baculovirus shuttle system. The recombinant plasmid DNA was then transfected into Sf6 insect cells. The permanent activation of the polyhedron gene promoter ensured increased local concentrations of the Cathepsin B protein. In all these situations, the experimental cells were different from the naturally living ones. Another mass spectrometry based method viz. fast photochemical oxidation of proteins (FPOP) has been explored by researchers in recent years to study protein conformational changes.22 This free radical mediated modification is fast and irreversible in nature. On the other hand, hydrogen/deuterium exchange is a reversible phenomenon where the rate of back exchange reaction is always minimized by reducing pH and temperature. In H/DX, the peptide backbone amide hydrogens contributed by 19 out of 20 naturally occurring amino acids, excluding proline, can be monitored. Thus conformation of any protein can be studied irrespective of its amino acid sequence. However in FPOP method, specific residues like cysteine and methionine are preferentially oxidized, limiting the applicability of the method to selective protein sequences.22 It has been reported that prolonged exposure of H2O2 is toxic to biological cells.23 As deuterium is an isotope of hydrogen, D2O is expected to be non-toxic to cells and the replacement of hydrogen by deuterium in proteins does not change its structure and chemical properties. The very high dynamic range of protein concentrations in a complex cellular environment and the variable ionization probabilities of different proteolyic peptides are the critical challenges to overcome for the successful implementation of the described method for other proteins in vivo. To visualize low abundant proteins in presence of high abundant ones and proteins with comparable abundance in a mixture, liquid chromatography based prefractionation at low temperature and acidic pH, prior to mass spectrometric analysis might play a crucial role.24 Additionally, the chromatographic separation step might be useful to increase the sequence coverage of an experimental protein. Sometimes there might be overlaps between isotope exchanged envelopes of peptides with nearby m/z. Under such circumstances, ion mobility based mass spectrometry that separates molecular ions on the basis of their size, shape and charge, could be used as an additional dimension of fractionation.25,26 Thus, hydrophobicity based separation in chromatographic step and size, shape and charge based separation in ion mobility mass spectrometry might enable to resolve proteins in a complex cellular environment. The present in-cell H/DX-MS approach would be useful to explore the structural changes associated with ligand/substrate binding, post-translational modifications and folding of proteins inside a living cell.



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

CONCLUSIONS The permeability of D2O across cell membrane and subsequent isotope exchange of backbone amide hydrogens made it feasible to study the structural transition of hemoglobin upon oxygenation inside RBC. As molecular mass is used as a probe, using multi-dimensional fractionation techniques coupled to mass spectrometry, the present method might be applicable to any protein inside a cell irrespective of its size, location and structural complexity.


Page 10 of 18

Page 11 of 18

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


REFERENCES (1) Zhou, H. X. FEBS Lett. 2013, 587, 1053-1061. (2) Roulhac, P. L.; Powell, K. D.; Dhungana, S.; Weaver, K. D.; Mietzner, T. A.; Crumbliss, A. L.; Fitzgerald, M. C. Biochemistry. 2004, 43, 15767-15774. (3) Tang, L.; Hopper, E. D.; Tong, Y.; Sadowsky, J. D.; Peterson, K. J. Anal Chem. 2007, 79, 5869-5877. (4) Weis, D. D.; Engen, J. R.; Kass I. J. J Am Soc Mass Spectrom. 2006, 17, 1700-1703. (5) Zhang, Z.; Smith, D. L. Protein Sci. 1993, 2, 522-531. (6) Kemmer, G.; Keller, S. Nat Protoc. 2010, 5, 267-281. (7) Davis, D. G.; Lindstrom, T. R.; Mock, N. H.; Baldassare, J. J.; Charache, S.; Jones, R. T.; Ho, C. J Mol Biol. 1971, 60, 101-111. (8) Perutz, M. F. Nature. 1970, 228, 726-739. (9) Paoli, M.; Liddington, R.; Tame, J.; Wilkinson, A.; Dodson, G. J Mol Biol. 1996, 256, 775-792. (10) Baldwin, J.; Chothia C. J Mol Biol. 1979, 129, 175-220. (11) Antonini, E.; Brunori, M. J Biol Chem. 1969, 244, 3909-3912. (12) Silva, M. M.; Rogers, P. H.; Arnone, A. J Biol Chem. 1992, 267, 17248-17256. (13) Devlin, T. M. In Textbook of Biochemistry with Clinical Correlations, 6th ed.; WileyLiss 2006, p351. (14) Perutz, M. F. Nature. 1970, 228, 734-739. (15) Shaanan B. J Mol Biol. 1983, 171, 31-59. (16) Mihailescu, M. R.; Russu, I. M. Proc Natl Acad Sci USA. 2001, 98, 3773-3777. (17) Fermi, G. J Mol Biol. 1975, 97, 237-256. (18) Ghaemmaghami, S.; Oas, T. G. Nat Struct Biol. 2001, 8, 879-882. (19) Sakakibara, D.; Sasaki, A.; Ikeya, T.; Hamatsu, J.; Hanashima, T.; Mishima, M.; Yoshimasu, M.; Hayashi, N.; Mikawa, T.; Wälchli, M.; Smith, B. O.; Shirakawa, M.; Güntert, P.; Ito, Y. Nature. 2009, 458, 102-105. (20) Ignatova, Z.; Gierasch, L. M. Proc Natl Acad Sci USA. 2004, 101, 523-528. (21) Koopmann, R.; Cupelli, K.; Redecke, L.; Nass, K.; Deponte, D. P. et. al. Nat Methods. 2012, 9, 259-262. (22) Espino, J. A.; Mali, V. S.; Jones, L. M. Anal Chem. 2015, 87, 7971-7978. (23) Hou, R. C.; Wu, C. C.; Huang, J. R.; Chen, Y. S.; Jeng, K. C. Ann N Y Acad Sci. 2005, 1042, 279-285. (24) Georgescauld, F.; Popova, K.; Gupta, A. J.; Bracher, A.; Engen, J. R.; Hayer-Hartl, M.; Hartl F. U. Cell. 2014, 157, 922-934. (25) Hilderbrand, A. E.; Myung, S.; Barnes, C. A.; Clemmer, D. E. J Am Soc Mass Spectrom. 2003, 14, 1424-1436. (26) Iacob R. E.; Murphy J. P. III; Engen J. R. Rapid Commun Mass Spectrom. 2008, 22, 2898-2904.



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

Acknowledgments: We acknowledge volunteers who provided blood sample for the study. Department of Science and Technology, Govt. of India is acknowledged for funding mass spectrometry facility at St. John’s Research Institute, Bangalore. M.M is supported by SRF fellowship from CSIR, Govt. of India. Dr. Anura V. Kurpad helped us in data processing and manuscript preparation. Dr. K. Srinivasan is acknowledged for providing helpful suggestions during manuscript preparation. We acknowledge Dr. Siddhartha Roy for valuable comments in manuscript preparation. We acknowledge Ms. Jennifer Pinto for technical assistance. Conflict of interest: The authors declare no conflict of interest. Author contributions: S.N performed the experiment, analyzed, interpreted the data and wrote the manuscript. G.M built the model, analyzed, interpreted the data and wrote the manuscript. M.M recorded MS data and analyzed it. B.M performed the experiment. A.K.M designed the entire research project, interpreted the model and the MS data and wrote the manuscript. Supporting Information: The envelopes for deuterium exchange kinetics of 13 peptides of deoxy and oxy hemoglobin are provided.


Page 12 of 18

Page 13 of 18

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


FIGURE CAPTIONS FIGURE 1. Schematic representation of in-cell H/DX analysis. Scheme shows in vivo H/DX of hemoglobin followed by cell lysis, proteolysis and subsequent mass spectrometry based analysis of isotope exchange kinetics. FIGURE 2. H/DX kinetics of peptide 1921.0 m/z. Panels (A) and (B) represent MALDI mass spectra for the peptide with m/z 1921.0 obtained on H/DX kinetics in deoxy and oxy states respectively. Panel (C) represents the best fit curve of exchange kinetics obtained from deoxyHb and oxy-Hb. The Y-axis is labeled with the number of deuterium atom incorporated and the X-axis with the corresponding exchange time. The inset shows m/z, globin subunit (β) and stretch of the peptic peptide of β globin chain. The error bars represent the standard deviation of 4 replicates of the experimental sets. FIGURE 3. H/DX kinetics of thirteen peptic peptides of deoxy-Hb and oxy-Hb. Panels A-M represent the best fit curves of exchange kinetics for the respective peptides obtained from deoxy-Hb and oxy-Hb. The Y-axis is labeled with the number of deuterium atom incorporated and the X-axis with the corresponding exchange time. m/z, globin subunits (α, β) and stretch of the individual peptides are indicated in each panel. The error bars represent the standard deviation of 4 replicates of the experimental sets.



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

Figure 1


Page 14 of 18

Page 15 of 18

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


Figure 2



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

Figure 3


Page 16 of 18

Page 17 of 18


1 2 Table 1. Analysis of H/DX kinetic parameters of 14 peptic peptides of deoxy-Hb and oxy-Hb 3 Deoxy Oxy N 4 Peptides ∑{(kiP i)oxy- (kiP i)doxy} (m/z, Sequence 5 k1×PA k2×PB k3×PC SSR k1×PA k2×PB k3×PC SSR i =1 6 Residues) 7 YQKVVAG 1869.1 VANALAH 7.48 0.73 2.88E-05 0.07 8.91 0.37 6.31E-04 0.02 1.07 8 β130-146 KYH 9 10 1494.9 VHLTPEEK 20.75 1.00 3.42E-05 0.01 23.44 2.36 1.35E-04 0.02 4.05 11 β1-14 SAVTAL 12 LSFPTTKT 13 1585.9 17.49 0.51 6.77E-05 0.12 34.53 0.42 1.32E-03 0.30 16.95 YFPHF 14 α34-46 ATLSELHC 15 1921.0 DKLHVDP 32.48 1.33 4.92E-02 0.06 58.61 3.03 3.54E-02 0.04 27.82 16 β86-102 EN 17 LVVYPWT 18 1308.7 3.09 0.47 7.32E-05 0.02 33.59 1.38 3.01E-03 0.02 31.42 β 32-41 QRF 19 20 1161.7 LVVYPWT 1.92 0.14 2.05E-02 0.15 54.56 1.15 4.42E-02 0.02 53.67 21 β 32-40 QR 22 AAHLPAEF 23 3428.0 TPAVHASL 16.54 0.68 4.78E-05 0.02 14.04 0.53 9.22E-02 0.01 -2.55 24 α110-141 DKFLASVS 25 TVLTSKYR 26 931.6 FRLLGNVL 4.81 0.22 2.62E-04 0.03 0.56 0.25 1.60E-04 0.05 -4.22 27 β103-110 28 WGKVNVD 29 1799.0 EVGGEAL 23.61 1.28 7.79E-05 0.03 9.92 1.70 6.09E-03 0.01 -13.26 30 β15-31 GRL 31 32 VLSPADKT 33 2910.6 NVKAAWG 29.61 0.76 1.70E-02 0.03 15.98 1.04 8.73E-04 0.02 -13.37 KVGAHAG 34 α1-29 EYGAEAL 35 36 1000.6 EVGGEAL 29.24 0.78 4.68E-05 0.02 13.71 1.47 3.44E-05 0.02 -14.84 37 β 22-31 GRL 38 39 VLSPADKT NVKAAWG 40 3326.8 KVGAHAG 57.78 1.11 5.08E-05 0.01 10.68 0.71 6.95E-05 0.01 -47.51 41 α1-32 EYGAEAL 42 ERM 43 44 VLSPADKT NVKAAWG 45 3473.9 KVGAHAG 112.49 1.06 1.72E-03 0.02 14.36 0.64 3.04E-03 0.03 -98.54 46 α1-33 EYGAEAL 47 ERMF 48 49 VCVLAHH 2051.1 FGKEFTPP 8.43 0.64 7.45E-04 0.02 7.35 0.69 2.06E-05 0.03 -1.04 50 β111-129 VQAA 51 52 53 P , P , and P are the number of fast, intermediate, and slow exchanging amide hydrogens with rate A B C 54 constants k , k 1 2, and k3, respectively. 55 56 57 58 59 60 ACS Paragon Plus Environment

Inference (deoxy → oxy) Flexible

Flexible Flexible Flexible Flexible Flexible

Rigid

Rigid

Rigid

Rigid

Rigid

Rigid

Rigid

Rigid


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

For TOC Only


Page 18 of 18

Protein StructureâFunction Correlation in Living Human Red Blood

Recommend Documents

Protein StructureâFunction Correlation in Living Human Red Blood