Subscriber access provided by Bibliothèque de l'Université Paris-Sud
Article
EXPLOITING HIS-TAGS FOR ABSOLUTE QUANTITATION OF EXOGENOUS RECOMBINANT PROTEINS IN BIOLOGICAL MATRICES: RUTHENIUM AS A PROTEIN TRACER Chengfeng Ren, Cedric E Bobst, and Igor A. Kaltashov Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.9b00504 • Publication Date (Web): 14 May 2019 Downloaded from http://pubs.acs.org on May 14, 2019
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 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 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.
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 17 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
EXPLOITING HIS-TAGS FOR ABSOLUTE QUANTITATION OF EXOGENOUS RECOMBINANT PROTEINS IN BIOLOGICAL MATRICES: RUTHENIUM AS A PROTEIN TRACER Chengfeng Ren,§ Cedric E. Bobst and Igor A. Kaltashov* Department of Chemistry, University of Massachusetts-Amherst, Amherst, MA 01003 ABSTRACT: Metal labeling and ICP MS detection offer an alternative to commonly accepted techniques that are currently used to quantitate exogenous proteins in vivo, but modifying the protein surface with metal-containing groups inevitably changes its biophysical properties and is likely to affect trafficking and biodistribution. The approach explored in this work takes advantage of the presence of hexa-histidine tags in many recombinant proteins, which have high affinity towards a range of metals. While many divalent metals bind to poly-histidine sequences reversibly, oxidation of imidazole-bound CoII or RuII is known to result in a dramatic increase of the binding strength. In order to evaluate the feasibility of using imidazole-bound metal oxidation as a means of attaching permanent tags to poly-histidine segments, a synthetic peptide YPDFEDYWMKHHHHHH was used as a model. RuII can be oxidized under ambient (aerobic) conditions, allowing any oxidation damage to the peptide beyond the metal-binding site to be avoided. The resulting peptide-RuIII complex is very stable, with the single hexa-histidine segment capable of accommodating up to three metal ions. Localization of RuIII within the hexa-histidine segment of the peptide was confirmed by tandem mass spectrometry. The RuIII/peptide binding appears to be irreversible, with both low- and high-molecular weight biologically relevant scavengers failing to strip the metal from the peptide. Application of this protocol to labeling a recombinant form of the 80 kDa protein transferrin allowed RuIII to be selectively placed within the His-tag segment. The metal label remained stable in the presence of ubiquitous scavengers and did not interfere with the receptor binding, while allowing the protein to be readily detected in serum at sub-nM concentrations. The results of this work suggest that ruthenium lends itself as an ideal metal tag for selective labeling of His-tag containing recombinant proteins to enable their sensitive detection and quantitation with ICP MS.
Quantitation of protein therapeutics is central to pharmacokinetic studies of biopharmaceutical products. Traditionally, this task was accomplished by immunoassays,1 but their limitations became apparent in the past decade,2-4 prompting a search for alternatives.5, 6 Mass spectrometry (MS) emerged in recent years as a powerful alternative to traditional immunoassays in dealing with multiple problems that complicate the use of the latter, such as the influence of autoand anti-reagent antibodies on the measurement outcome, as well as the lack of concordance across platforms.5 MS-based quantitation is typically done at the peptide level, with proteolytically produced fragments analyzed by LC/MS or LC/MS/MS.7 The benefits provided by protein quantitation at the peptide level are two-fold: first, it allows immunoaffinity capture at the peptide level to enhance selectivity and eliminate the influence of auto-antibodies. Second, it enables relatively straightforward quantitation via stable isotope labeling with internal standards introduced either at the peptide8 or the whole protein9 levels. However, reliance on a single (or even multiple) proteolytic fragments as a means of protein detection has its own limitations,5, 10 the most significant of which is the inability to differentiate between two or more proteoforms with nearidentical amino acid sequences (if the difference between them resides outside of the polypeptide segments used for quantitation). Above and beyond the frequently cited inability of peptide-based quantitation to make a distinction among the proteoforms representing various post-translational modifications (the result of the protein drug biotransformation),
the presence of structurally similar endogenous proteins may also interfere with the quantitation process. The latter would become particularly difficult to discriminate against if the protein drugs (the exogenous species) are designed after specific endogenous proteins. For example, targeted drug delivery may be achieved by using endogenous transport proteins as blueprints for designing delivery vehicles, including transferrin,11 ferritin,12 and albumin13 to name just a few. In all of these cases, the presence of abundant endogenous proteins in biological fluids would make the quantitation of the exogenous protein with a nearly-identical amino acid sequence extremely difficult when relying on one or few proteolytic fragments. An alternative to peptide-based quantitation is offered by intact protein detection, an approach that is becoming increasingly popular when a distinction among various proteoforms must be made. However, it relies on affinity capture10, 14, 15 and offers significantly lower sensitivity compared to the peptide-level quantitation protocols. An orthogonal approach to protein quantitation is offered by inductively-coupled plasma (ICP) MS, a technique that is steadily gaining recognition as a powerful bioanalytical tool.16 Although its applications in quantitative drug analyses were previously limited to platinum-containing small-molecule medicines, growing demands of proteomics and clinical analysis have stimulated interest in exploring the utility of ICP MS as a tool for absolute protein quantitation.17, 18 For example, sensitive detection of proteins and peptides in complex biological matrices can be achieved by labeling them with metal
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
tags using a variety of strategies.19 Unfortunately, in most cases labeling involves chemical modification of the protein with either organomercurials or metal chelate reagents. This inevitably affects a range of protein properties and may alter its behavior in vivo, making the conclusions of biodistribution studies relying on such tags highly suspect. The problem of altering the protein structure can be avoided by using ICP MSbased immunoassays (where the protein detection is enabled by a metal-tagged antibody);17, 20, 21 however, this approach is likely to fail in drawing a distinction among structurally similar/nearly identical proteins (e.g., an exogenous protein therapeutic and its endogenous counterpart that was used as a blueprint in designing the protein drug). Problems associated with chemical labeling can also be avoided if the ICP MS detection relies on “natural” tags, such as sulfur;20 several metal ions have also been used as cognate tags for a range of metalloproteins.22 However, these strategies suffer from the lack of specificity due to the ubiquitous nature of the elements used as tags, which limits their applications. The specificity issue can be addressed for some metalloproteins by substituting the cognate metals with the ones that are not present in living organisms and therefore, can serve as unique tags for ICP MS detection. Thus, substituting Fe3+ with Ga3+ in methemoglobin may allow detection of exogenous hemoglobin/haptoglobin complexes with high sensitivity.23 Likewise, substituting Fe3+ with In3+ in transferrin not only enables sensitive detection of transferrin-based therapies in blood and cerebrospinal fluid, but also allows their biodistribution patterns to be determined within various organs of animal models.24 Unfortunately, this strategy is restricted to metalloproteins and its success hinges on identification of exogenous metals that can be used to replace the cognate ones without compromising the binding affinity and changing the properties of the protein.25 The majority of therapeutic proteins are not metalloproteins; therefore, a different strategy must be sought in order to enable their tagging with metals for ICP MS detection. One common feature of most recombinant proteins is the presence of a Histag, which is introduced into their amino acid sequences to facilitate isolation and clean-up. While there is a range of imidazole-binding metals, most of them bind to protein His-tags reversibly, making them unfit to serve as reliable tags for biodistribution quantitation by ICP MS. For example, the hexahistidine-bound Ni2+ ions appear to be labile and are readily stripped from the His-tags of recombinant proteins by albumin,24 a scavenger protein present in blood plasma at high concentrations. This renders any attempt to use nickel as a tracer of recombinant proteins in biological fluids and tissues meaningless. Reversibility of the poly-histidine/divalent metal binding is caused by the kinetic lability of such complexes, which are easily disturbed by chelators or pH change. However, more stable and kinetically inert complexes can be produced following metal oxidation (e.g., Co2+ to Co3+), which diminishes the ligand exchange rate by several orders of magnitude.26 Unfortunately, the formation of the Co3+ complexes is very slow, necessitating indirect preparation methods (e.g., by making the complex of the exchange-labile divalent metal first, followed by its oxidation).26 The inherent danger of this procedure when applied to a divalent metal/recombinant protein complex is the possibility of inducing oxidative damage beyond the His-tag, especially within the oxidation-prone structural elements of the protein (e.g., methionine side chains).
Page 2 of 17
An alternative approach to labeling His-tag segments of recombinant proteins with metals utilizes introduction of reactive amino acid residues that provide further stabilization to the protein/metal complex. For example, stable technetium/poly-histidine complexes can be readily produced,27 but optimal binding of both Tc28 and its next-of-kin Re29 requires introduction of a cysteine residue in close proximity to the hexa-histidine sequence. The need to introduce nonstandard His-tags containing chemically active amino acids reduces the scope of proteins amenable to quantitation using this strategy. The present study aims at identifying exogenous metals that can bind to polypeptides containing “standard” His-tags selectively and irreversibly without inflicting damage/structural changes beyond the hexa-histidine segment. Our work on producing stable metal/hexa-histidine complexes using cobalt had met with mixed success and will be reported elsewhere. Briefly, incorporation of Co2+ into a hexa-histidine segment of a model peptide followed by oxidation of the complex to produce a stable metal tag invariably led to significant oxidative damage beyond the metal-binding site. In contrast, ruthenium (III) appears to be an ideal metal tag, which not only can be readily incorporated into the hexa-histidine sequences with a high degree of selectivity, but also remains bound under a range of physiological conditions (including the mildly acidic environment of endosomes) even in the presence of metal scavengers. EXPERIMENTAL SECTION Materials. Model peptide YPDFEDYWMKHHHHHH was purchased from Biomatik (Wilmington, DE), purified using reversed-phase HPLC, and its amino acid sequence was verified by MS/MS. [RuII(NH3)5Cl]Cl, as well as human serum albumin, transferrin, ascorbate, glutathione and citrate were purchased from Millipore-Sigma (St. Louis, MO). The ectodomain of human transferrin receptor (TfR) and the recombinant form of human serum transferrin with intact His-tag (rhTf) were provided by Prof. Anne B. Mason (University of Vermont School of Medicine, Burlington, VT). Human serum was acquired from Equitech-Bio, Inc. (Kerrville, TX). All other chemicals and reagents used in this work were of analytical grade or higher. Metal/peptide complexes were formed by incubating the peptide (20 μM) with excess metal (50:1 molar ratio) in a series of buffers (50 mM Tris, pH 7.4; 40 mM Tris, 10 mM sodium acetate, 50 mM sodium bicarbonate, pH 7.4; 50 mM ammonium bicarbonate, pH 8.0) for 12-36 hrs.; binding equilibrium was typically reached by 24 hrs. at 37 oC. Hydrolysis of the metal/peptide complex was carried out by 20hr incubation with 50-molar excess of cyanogen bromide in 70% formic acid at 25 OC in the dark. Labeling of rhTf was carried out by incubating the protein solution in the 50 mM Tris buffer with a 20- to 50-molar excess of [RuII(NH3)5Cl]Cl at 37 OC for 48-120 hrs (as specified in the text). Liquid chromatography. The initial purification of the model peptide, as well as separations of the peptide/metal complexes were carried out using a 4.6 x 150 mm reversed phase column (Zorbax 300SB-C8, Agilent, Santa Clara, CA) on an Agilent 1200 (Agilent, Santa Clara, CA) HPLC. Mass spectrometry. Structural characterization of the peptide/metal and rhTf/metal complexes was carried out with a SolariX 7 (Bruker Daltonics, Billerica, MA) Fourier transform ion cyclotron resonance mass spectrometer equipped with a standard electrospray ionization source. Ion fragmentation was induced by electron capture dissociation (ECD) of trapped ions
ACS Paragon Plus Environment
Page 3 of 17 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
in the cyclotron cell using the following instrumental parameters: pulse length 0.6 s, bias 1.1 V, lens 10 V. Singly charged histidine clusters were used for mass calibration. Native MS measurements of Ru-labeled rhTf and TfR complex were carried out using a QStar XL (AB/Sciex, Toronto, Canada) hybrid quadrupole/time-of-flight mass spectrometer. Detection of ruthenium in serum spiked with Ru-labeled rhTf was carried out using NexION 300X (PerkinElmer, Waltham, MA) ICP mass spectrometer following its overnight digestion with 49% HNO3 and 7% H2O2 (by volume) at 37 OC. The kinetic energy discrimination (KED) mode was used to reduce polyatomic interferences. RESULTS AND DISCUSSION Ru(II) displays high affinity towards imidazole groups (by preferring histidine side chains over carboxylate and primary amine ligands30, 31), and the imidazole-bound metal can be readily oxidized (e.g., the standard reduction potential for the [(C3N2H4)2(NH3)4RuIII]3+/[(C3N2H4)2(NH3)4RuII]2+ transition is only 0.12-0.15 V, depending on the configuration32). In fact, aerobic conditions had been shown to be sufficient to drive the oxidation and produce ligand exchange-inert Ru(III)/imidazole complexes.33 Furthermore, ruthenium is an exogenous element and is present in organisms only as a result of poisoning or administration of Ru-based therapies;34 therefore, using this metal as a tag should enable detection of tagged recombinant proteins with much high sensitivity. It may, therefore, seem surprising that ruthenium has not been explored as a means of polypeptide tagging for sensitive ICP MS detection beyond producing organometallic conjugates at tryptophan side chains.35 In order to ensure that this metal can be used for robust and damage-free labeling of recombinant proteins containing hexahistidine segments, we adopted the following set of criteria. First, a mild oxidation route should be available for the Ru2+/polypeptide complex to produce a stable trivalent metal/polypeptide complex without inflicting oxidative damage beyond the hexa-histidine segment (a criterion we have not been able to meet when converting labile Co2+/hexa-histidine complexes to exchange-inert Co3+ complexes, vide supra). Second, in order to maximize the sensitivity of ICP MS measurements the yield of Ru3+/polypeptide complex formation should be relatively high, but all metal ions should be localized within the hexa-histidine segment of the polypeptide. Lastly, the complexes should be sufficiently stable to prevent the loss of metal tags under physiologically relevant conditions (e.g., at neutral pH in the presence of albumin and at endosomal pH in the presence of small organic chelators). These criteria were applied to study ruthenium interaction with a model peptide, YPDFEWFEMKHHHHHH. The latter was designed to incorporate a hexa-histidine segment in addition to a decapeptide sequence containing a methionine residue (introduced as an indicator of the occurrence of oxidative damage beyond the presumed metal-binding site). The decapeptide sequence also contains three closely spaced acidic residues that can potentially interact with metals, but are harder bases compared to imidazoles (which preferentially interact with Ru2+, a soft/borderline acid). These “metal baits” were introduced into the model peptide sequence to enable meaningful verification that the oxidized metals (which become harder acids compared to their reduced forms) are still localized within the hexa-histidine region despite the availability of alternative ligand sites.
Ruthenium(III)/peptide complexes can be produced at high yield without generating oxidative damage beyond the presumed metal-binding sites. Our initial attempts to produce RuIII/model peptide complex by incubating the peptide in the presence of significant molar excess of [(NH3)5RuIICl]Cl under anaerobic condition (to produce RuII/peptide complex) followed by brief air-induced oxidation resulted in a disappointingly modest yield of the desired product despite the large (50-fold) molar excess of the metal (see Figure S1 in Supporting Material). Indeed, even though the produced metal/peptide complexes contained the oxidized form of Ru, the majority of the peptide molecules remained metal-free. Furthermore, even though the presence of up to five NH3 ligands within the peptide/metal complexes was indicative of only one (out of six) imidazole group participating in RuIII coordination, no complexes of 1:2 or higher stoichiometry were detected. These initial Ru-labeling experiments were carried out in a variety of buffers, with 50 mM ammonium bicarbonate giving the highest yields (20%), followed by Tris (
