Fingerprinting Protein Populations with Post ... - ACS Publications

Subscriber access provided by TUFTS UNIV

Article

An environment-sensitive “turn-on” fluorescent polyamino acid: Fingerprinting protein populations with post-translational modifications Shunsuke Tomita, Sayaka Ishihara, and Ryoji Kurita ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.7b05360 • Publication Date (Web): 19 Jun 2017 Downloaded from http://pubs.acs.org on June 21, 2017

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.

ACS Applied Materials & Interfaces 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 32

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

ACS Applied Materials & Interfaces

An environment-sensitive “turn-on” fluorescent polyamino acid: Fingerprinting protein populations with post-translational modifications Shunsuke Tomita*, Sayaka Ishihara, and Ryoji Kurita* Biomedical Research Institute, National Institute of Advanced Industrial Science and Technology, and DAILAB, 1-1-1 Higashi, Tsukuba, Ibaraki 305-8566, Japan. Keywords: proteins, post-translational modifications, multivariate analysis, sensor array, polyelectrolytes.

ACS Paragon Plus Environment

1


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 2 of 32

Abstract

The identification of post-translational modifications (PTMs) in proteins has been of particular interest in the elucidation of human diseases and the improvement of therapeutic proteins. Herein, we report a novel strategy toward the construction of fingerprint-based PTMsensing systems as an alternative to conventional specific recognition tools. Our strategy is based on poly-L-lysine (PLL) derivatives with two distinct properties suitable to fingerprint-based protein-sensing: i) a “turn-on” fluorescent signal upon binding to proteins, and ii) conditiondependent cross-reactivity toward proteins and PTMs. One type of PLL derivative under varying solution properties (pH value and ionic strength) was sufficient to construct a sensing array that produces unique fluorescence fingerprints for structurally similar mammalian albumins with/without a wide variety of chemical modifications corresponding to PTMs. This approach was also applicable for the recognition of deviations in physicochemical properties of proteins as a result of realistic glycation and phosphorylation events. Multivariate analyses of the thus obtained fingerprints were able to successfully identify analytes with 100% accuracy (qualitatively and quantitatively) in all cases. This study thus demonstrates for the first time a fingerprint-based sensing of proteins with/without PTMs using a single, highly accessible, and tunable synthetic polymer, and accordingly offers a powerful platform for simple highthroughput sensing of PTMs in proteins.


2

Page 3 of 32

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


Introduction Most proteins are modified after their translation in vivo, a process that is referred to as posttranslational modification (PTM). In recent years, the detection of PTMs has been recognized as a key aspect in the elucidation of human diseases,1-3 as well as in the improvement of therapeutic proteins.4-6 Therefore, much of the current effort in PTM research has been devoted to the development of methodologies to identify these modifications. The most widespread tool for the easy and practical tracking of PTMs is the use of PTM-specific antibodies.7,8 However, commercially available antibodies commonly suffer from instability, high costs, and often lack specificity toward a targeted PTM.9 In addition, most available PTM-specific antibodies are polyclonal and thus represent exhaustible resources, while different batches show distinct profiles.10,11 In addition to antibody-based techniques, the use of selective synthetic receptors to detect PTM events has been extensively investigated.12-14 Even though recognition of PTMs has been achieved by these receptors, it is necessary to synthesize highly selective receptors for each protein with PTMs. Thus, the development of an alternative strategy to construct versatile PTM detection systems without relying on specific recognition pairs remains essential. Fingerprint-based sensing, which exploits the pattern-recognition of unique optical fingerprints for individual analytes, is a promising prospective alternative. Such fingerprints are acquired using arrays of “cross-reactive” molecules that can interact in different ways with target analytes. This sensing strategy has been used to discriminate drug-related molecules,15-18 proteins,19-25 and mammalian cells.26-29 We have previously developed arrays of cross-reactive polyion complexes between enzymes and charged polymers for the identification of proteins,30-32 real biofluids such as mammalian sera,32,33 and cell culture media containing proteins secreted from mammalian cells.34


3


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 32

Fingerprint-based sensing systems can be tuned to target analytes by selecting a combination of suitable elements, allowing the identification of challenging analytes with similar characteristics. Such systems have been applied to detect PTMs in peptides, such as tripeptides and their phosphorylated analogues35 as well as methylated, acetylated, and phosphorylated histone tail peptides.36 These PTMs have been discriminated by fluorescent dyes bound to and quenched by synthetic host molecules. The target peptides interact with the synthetic molecules and displace the fluorescent dyes, thus generating a signal in the so-called indicator displacement assay. It is moreover possible to detect phosphorylation events in protein kinases based on such indicator displacement assays using synthetic molecules conjugated to probe peptides bearing specific affinity for phosphorylated proteins.37 However, previously reported fingerprint-based PTM-sensing systems are based on supramolecular conjugates of highly tailored molecules, which limits the operability and the versatility of their applications. To create simpler and more practical systems, it is desirable that the molecules themselves exhibit: i) a response with high signal-to-noise ratio (e.g. turn-on mode)38,39 and ii) cross-reactivity toward PTM events. Herein, we report a new system based on poly-L-lysine (PLL) derivatives that incorporates environment-sensitive fluorophores (Figure 1A), which are capable of generating a fluorescence signal in a turn-on manner upon binding to proteins (Figure 1B). The cross-reactivity of the PLLs toward proteins with/without PTMs was found to depend strongly on both the pH value and the ionic strength of the aqueous solution. As a result, a sensing array consisting of a single PLL derivative

produced

unique

fluorescence

fingerprints

(Figure

1C)

that

reflect

the

physicochemical properties of i) structurally similar mammalian albumins, ii) a wide variety of chemical modifications corresponding to PTMs, and iii) realistic glycation and phosphorylation


4

Page 5 of 32

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


events. Multivariate analyses of the fingerprints are able to reliably discriminate these proteins and PTMs both qualitatively and quantitatively.

Figure 1. (A) Chemical structures of cationic poly-L-lysines (PLLs) modified with environment-sensitive fluorophores. (B) Schematic representation of fluorogenic interactions with a protein with/without PTMs. (C) Resulting fluorescence fingerprints reflecting each protein obtained under different solution conditions.

Results and discussion PTMs usually introduce new functional groups into amino acids (Figures 2), which results in slight modulations of the physicochemical properties of the protein surfaces, including their net


5


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 6 of 32

charge and hydrophobicity. For this study, we chose cationic poly-L-lysine (PLL) for the recognition of such PTM-induced modulations. The lysine side chains bear positively charged amino (pKa ~ 10.5) and hydrophobic n-butyl groups. We envisioned that PLL should bind differently to proteins containing different PTMs via multiple electrostatic and hydrophobic interactions, whose cross-reactivity should be suitable for fingerprint-based sensing. In order to generate fluorescent responses reflecting interactions between PLL and proteins with PTMs in a turn-on

manner,

we

introduced

the

environment-sensitive

fluorophores

5-

(dimethylamino)naphthalene-1-sulfonyl chloride (Dnc-Cl) and 4-(N,N-dimethylaminosulfonyl)7-fluoro-2,1,3-benzoxadiazole (DBD-F) into the terminal amino groups of the polymers (Figure 1A). In polar aqueous solutions, these fluorophores exhibit weak fluorescence, which intensifies sharply upon decreasing the polarity of the microenvironment,40,41 e.g. upon protein binding.42-44 Almost the same number of terminal amino groups of PLL were modified with Dnc-Cl (PLLDnc) and DBD-F (PLL-DBD) (Figure 1A; for details, see Section 1 of the Supporting Information). Initially, a fluorescence titration of structurally similar mammalian serum albumins (bovine: BSA, human: HSA, equine: ESA, and rabbit: RSA; Figure 2A) was carried out in order to examine whether the modified PLLs displayed turn-on responses and recognized the structural differences in these albumins (Figures 3A and S1). As expected, PLL-Dnc exhibited a weak emission at pH = 5.5, which experienced a blue shift (∆λ = 23 nm) and a sevenfold increase in fluorescence intensity upon addition of ESA. Although similar responses were observed for PLL-DBD, the fourfold increase in emission intensity was lower than that for PLL-Dnc, indicating less suitability for the sensitive and precise discrimination of proteins (for a detailed comparison of PLL-DBD and PLL-Dnc, see Section 3 of the Supporting Information). In contrast, DBD-conjugated low-molecular weight probes for selective protein detection have


6

Page 7 of 32

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


previously shown a markedly higher enhanced fluorescence than those involving Dnc groups.44 The origin of this behavior remains to be determined precisely. However, the amino acid environment around the binding site probably engages in relatively non-specific interactions with PLL, which should favor the enhancement of the Dnc fluorescence. Consequently, PLL-Dnc was selected for further sensing experiments.

Figure 2. Profiles of analyte proteins used in this study: (A) mammalian albumins; (B) bovine and human serum albumins with various chemical modifications corresponding to PTMs. For examples of these PTMs in organisms, see Table S1; (C) inactivated and activated GST-tagged human ERK1; Φsurface values of proteins were calculated based on the Miyazawa–Jernigan hydrophobicity scale.45


7


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 8 of 32

Figure 3. Characterization of PLL-Dnc: (A) fluorescence spectra of 5.0 µg/mL PLLDnc upon addition of ESA in 18 mM MES (pH 5.5, λex = 340 nm, [ESA] = 0–59 µg/mL); (B-D) binding isotherms for 5.0 µg/mL PLL-Dnc upon addition of four mammalian serum albumins in (B) 18 mM MES (pH 5.5), (C) 18 mM MES (pH 5.5) + 25 mM NaCl, and (D) 18 mM EPPS (pH 8.5); λex = 340 nm, λem = 510 nm.

It is possible that the cross-reactivity of cationic PLL-Dnc depends on certain solution properties such as the pH value and ionic strength. As the isoelectric point (pI) of proteins used in this study is between 4.8 and 6.1 (Figure 2), the net charge of such proteins alters remarkably at pH = 5.5-8.5. In addition, the contribution of the hydrophobicity to the interaction between PLL-Dnc and the proteins increases due to electrostatic screening upon increasing the ionic strength of the solution. Figures 3B-D show the titration isotherms corresponding to the fluorescence increase of PLL-Dnc under different solution conditions. Evidently, PLL-Dnc interacts differently with each albumin, and the responses against albumins are markedly influenced by both the pH value and the ionic strength of the solution. For instance, ESA displays the highest response toward PLL-Dnc at pH = 5.5 (Figure 3B), while the ESA response is lower than that of HSA and BSA in the presence of 25 mM NaCl (Figure 3C). At pH = 8.5, all


8

Page 9 of 32

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


albumins display a similar behavior, yet the fluorescence changes as a response to the addition of ESA and RSA are slightly different from those in response to the addition of BSA and HSA (Figure 3D). The pI of these serum albumins is at pH = ~5.6 (Figure 2A), i.e., hydrophobic interactions should dominate close to the pI (pH = 5.5). Conversely, all albumins are negatively charged at pH > 5.5, presumably resulting in similar electrostatically driven affinities toward PLL-Dnc at pH = 8.5. In addition to the maximum fluorescence intensities, the dissociation constants (Kd) and the number of binding sites (n) also depend on the nature of the serum albumins (for details, see Section 3 of the Supporting Information). Dynamic light scattering (DLS) and far-UV circular dichroism (CD) measurements suggested that the addition of PLL-Dnc caused the formation of aggregate-like complexes and partial denaturation of ESA (for details, see Section 3 of the Supporting Information). Overall, the modulation of the cross-reactivity of PLL-Dnc toward these serum albumins was achieved by altering the characteristics of the aqueous solution, despite the high sequence identity (~70%) and very close resemblance of the tertiary structure of these serum albumins.46 The discriminative power of PLL-Dnc is thus not only the result of the cationic amino groups and the hydrophobic n-butyl groups of the lysine side chains, but also of the introduced dansyl groups themselves. Encouraged by the easily controllable cross-reactivity of PLL-Dnc toward structurally similar albumins, we decided to attempt fingerprint-based PTM sensing. In addition to the four aforementioned serum albumins and two more mammalian albumins (chicken ovalbumin: OVA and bovine lactalbumin: BLA), we initially chose albumins with chemical modifications


9


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 10 of 32

corresponding to PTMs: phosphorylation (BSA-PS and BSA-PY),20 acetylation (BSA-Ac), methylation (BSA-Me), and glycation (BSA-Lac and HSA-Glc) (Figure 2B). For the sensing procedure, each protein solution was mixed with PLL-Dnc (5.0 µg/mL) to a final protein concentration of 20 µg/mL in six different buffer solutions, i.e., 18 mM 2-(Nmorpholino)ethanesulfonic acid (MES) at pH = 5.5, 18 mM 3-(N-morpholino)propanesulfonic acid (MOPS) at pH = 7.0, and 18 mM 3-[4-(2-hydroxyethyl)-1-piperazinyl]propanesulfonic acid (EPPS) at pH = 8.5 in the presence/absence of 25 mM NaCl on a 96-well plate. The fluorescence signals from each protein/sensor element combination were recorded as (I-I0) at four different channels [λex/λem: 340/480 (Ch1), 340/520 (Ch2), 340/560 (Ch3), and 320/520 (Ch4)], generating 6 replicates × 6 solutions × 4 channels × 13 analytes including a blank solution. The responses are summarized visually as a heat plot in Figure 4, which is a useful way to highlight the rich cross-reactivity of this system47 (for raw data, see Table S2 and Figure S2). The obtained fingerprints are highly reproducible, presumably due to the simplicity of the sensing format and procedure. It is evident that phosphorylation and acetylation markedly increase the fluorescence. As shown in Figure 2B, both PTMs enhance the net negative charge of the protein; phosphorylation adds a -2 charge to the side chains of Ser and Tyr residues, while acetylation neutralizes the positively charged amino groups of Lys residues. The increase in net negative charge of BSA strengthens the interactions with the cationic PLL-Dnc. In contrast, methylation on the carboxylate side chains of Glu and Asp neutralizes the negative charge (Figure 2B), thus decreasing the affinity toward PLL-Dnc.


10

Page 11 of 32

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 4. Heat plot of the fluorescence fingerprints of 20 µg/mL albumins with/without chemical modifications obtained from the PLL-Dnc system. Six replicates are shown for each analyte.

The abundance of discriminative information was statistically evaluated using linear discriminant analysis (LDA), a pattern recognition algorithm that provides a graphical output that offers insight into the clustering of the data and information on the classification ability.48 Surprisingly, the linear discriminant score plot revealed 13 well-separated clusters corresponding to individual proteins (Figure 5; for two dimensional plots, see Figure S3), affording 100%

Figure 5. Discriminant score plot for 20 µg/mL albumins with/without chemical modifications corresponding to PTMs obtained from the PLL-Dnc system.


11


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 12 of 32

classification accuracy in a leave-one-out procedure (jack-knife classification).49 A single PLLDnc sensor is thus capable of separating four mammalian serum albumins (BSA, HSA, RSA, and ESA), although the discrimination of such homologous proteins previously required welldesigned multiple reactive fluorophores50 or cross-reactive conjugates between, e.g., a graphene oxide and DNA aptamers selected by SELEX,51 or an enzyme and block-copolymers.31 Among all 13 analytes, these four serum albumins were clustered (dashed circles in Figure 5), indicating that the similarity between serum albumins is recognized by the PLL-Dnc system. More importantly, chemical modifications in albumins caused large changes in their cluster position. The clusters corresponding to phosphorylation (BSA-PS and BSA-PY) and glycation (BSA-Lac and HSA-Glc) were found in close proximity with respect to each other, and well separated from the other modifications (methylation and acetylation). It should be noted here that HSA and HSA-Glc could also be successfully discriminated, even though only two glycosyl groups were introduced into HSA (Table S1). Furthermore, another 39 protein samples were prepared and used as unknowns in a blind test. These unknowns were classified according to their Mahalanobis distances to each group, achieving 100% accuracy (39/39 samples, Table S3). Similar results were obtained from an unsupervised hierarchical clustering analysis (HCA).48 In contrast to LDA, the dimensions of the datasets are not reduced in HCA, i.e., the calculated distances between clusters correspond to similarities in the fingerprints of the analytes. Thirteen analyte clusters were distinguished in the corresponding dendrogram, wherein phosphorylated and glycated albumins were individually clustered (Figure S4). Hence, the PLL-Dnc system has considerable power for the identification of modifications introduced into proteins. We then investigated which elements are related and contributed to the discrimination in order to improve the sensor design. Each data element was initially divided by the root mean square of


12

Page 13 of 32

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


the corresponding data element set for normalization,32 prior to being analyzed using HCA (Figure 6). Channels with the same solutions were initially clustered (Figure 6A), indicating that the use of different λex/λem combinations plays an insignificant role when generating differential fingerprints compared to that of the ionic strength and the pH value. The pH values of the solutions with/without NaCl were subsequently clustered (Figure 6B). However, the dendrogram shows clustering at 1.19 distance units (d.u.) for the elements with/without NaCl at pH = 5.5, while clustering for the elements at pH = 7.0 and 8.5 was observed at 0.67 d.u. (Figure 6C), suggesting that choosing a pH value for the solution that is close to the pI of the target protein and changing the ionic strength of the solution at this pH value represents an effective way to construct a fingerprint-based protein-sensing system. As previously mentioned, this is probably due to increases in the contribution of the hydrophobic interactions to the binding of albumins with PLL-Dnc by electrostatic screening. These interpretations are supported by a principal component analysis (PCA) (for details, see Section 4 in the Supporting Information).48


13


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 14 of 32

Figure 6. Clustering analysis of discriminative elements of the PLL-Dnc system. The hierarchical clustering dendrogram based on the Euclidean distance with the Ward method was created using a data set (24 solutions × 13 analytes × 6 replicates) divided by the root mean square of the corresponding solution data. Areas "A" and "B" represent clustering of the same solution properties and same pH values, respectively. "C" indicates the distance between the clustering point in the presence/absence of NaCl at pH = 5.5 and that at pH = 7.0/8.5.

After the successful qualitative discrimination of albumins with/without chemical modifications corresponding to PTMs, we explored if the PLL-Dnc system is able to quantify the population of modified proteins. To this end, we first selected HSA glycation, which is a nonenzymatic PTM that yields so-called Amadori products.52,53 HSA glycation is routinely used to monitor the glycemic control in diabetic patients.54 In healthy individuals, ~10% of HSA is glycated, while this amount increases by a factor of 2-5 in diabetic patients.53 Conventional


14

Page 15 of 32

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


detection methods are based on multistep assays including boronate affinity chromatography, immunoassay-related techniques, and enzymatic methods.53 Mixtures of various concentrations of HSA and HSA-Glc were prepared and analyzed (Table S4). The LDA of the PLL-Dnc system resulted in a clear separation of the clusters (Figure 7A) with a 100% correct classification in both the jack-knife procedure and the blind test (24/24 samples, Table S5). In addition, the analysis yielded a smooth trend for the HSA:HSA-Glc ratio for a fixed total concentration of 30 µg/mL. Most importantly, the clusters from 0% HSA-Glc (30/0; 30 µg/mL HSA) to 33% (20/10; 20 µg/mL HSA and 10 µg/mL HSA-Glc) were well separated, indicating the ability of our approach to detect traces of HSA-Glc at the diagnostic level. We then attempted to determine the concentration of non-phosphorylated and phosphorylated albumins. As phosphorylation controls key cellular processes, such as the cell cycle, circadian rhythm, and apoptosis,55 it is the most extensively studied among all identified PTMs. As for the glycation events, the LDA yielded a clear discrimination of different concentrations of BSA/BSA-PS mixtures with 100% accuracy (Figure 7B, Tables S6 and S7). BSA clusters moved along the y-axis with increasing concentration, while movement along the x-axis was observed for BSA-PS, suggesting the recognition of unique changes in the BSA characteristics as a result of phosphorylation.


15


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 16 of 32

Figure 7. Discriminant score plot for the quantitative response of (A) HSA and HSAGlc with a total concentration of 0–30 µg/mL, (B) BSA and BSA-PS with a total concentration of 0–9 µg/mL, (C) ERK1-Ac and ERK1-In with a total concentration of 0–0.6 µg/mL using the PLL-Dnc system. The 95% confidence ellipses for the individual analytes are shown.

In order to extend our fingerprint-based sensing, we envisioned that the PLL-Dnc system could be used to fingerprint and discriminate more realistic phosphorylation events. Therefore, ERK1 was chosen as the target analyte (Figure 2C). ERK1 is a mitogen-activated protein kinase (MAPK) that plays a central role in MAPK cascades and in the regulation of cell growth and differentiation.56 Since phosphorylation of both Thr202 and Tyr204 is required for the activation of ERK1,56 activated (ERK1-Ac) and inactivate (ERK1-In) ERK1 should be distinguished if the changes in the physicochemical characteristics of ERK1 induced by the two phosphorylation events are recognized by the PLL-Dnc system. The responses obtained for ERK1-Ac/ERK1-In mixtures were analyzed using selected conditions (see Table S8) and, again, a 100% correct classification was accomplished at concentrations as low as 0.6 µg/mL (~9.0 nM) (Figure 7C).


16

Page 17 of 32

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


These results demonstrate the potential of our strategy for the in vitro analysis of populations of proteins with PTMs in a quantitative manner. It should be noted that specificity was not considered when constructing the sensing arrays, which stands in stark contrast to previously reported fingerprint-based sensing techniques for protein PTMs, as these use peptides that exhibit specific affinities to phosphorylated proteins.37

Conclusions In summary, we have developed a fingerprint-based system applicable to PTM sensing using a PLL derivative that is partially modified with environment-sensitive fluorophores (PLL-Dnc). The binding of proteins with PLL-Dnc generates turn-on fluorescence signals, and the crossreactivity toward the proteins depends significantly on the solution properties (pH value and ionic strength). Therefore, the use of sensing arrays consisting of a single PLL-Dnc is able to successfully discriminate qualitatively and quantitatively not only between structurally similar mammalian albumins, but also between PTMs introduced into albumins and kinases. To the best of our knowledge, our system shows, for the first time, the discrimination of proteins with/without PTMs by fingerprint-based sensing using one type of synthetic polymer. Although PLL-Dnc exhibits high discriminatory power, it can be easily prepared, and its properties are suitable to construct a fingerprint-based protein-sensing system with a minimal number of sensor elements. As this polymer displays potential for further tunability, it offers a powerful platform for both sensing of PTMs and other physicochemical changes in proteins. Each PTM affords a particular fingerprint, and thus, it is expected that our approach will aid the identification and discrimination of multiple types of PTMs in a single protein or in protein


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

Page 18 of 32

mixtures, making it a useful tool to gain insight into the aberrant PTM patterns of human diseases1-3 and the quality of protein-based biopharmaceuticals.4-6

ASSOCIATED CONTENT Supporting Information. Experimental details, response profiles, statistical analyses, fluorescent spectra, detailed properties of the polymers, Figures S1-S10 and Tables S1-S10.

AUTHOR INFORMATION Corresponding Authors *E-mail: [email protected]; [email protected] Notes The authors declare no competing financial interest.

ACKNOWLEDGMENT We thank Dr Yoshio Suzuki (Health Research Institute, National Institute of Advanced Industrial Science and Technology) for help with the polymer synthesis and Prof. Dr Kentaro Shiraki (Faculty of Pure and Applied Sciences, University of Tsukuba) for help with the titration


18

Page 19 of 32

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


experiments. This work was supported by JSPS KAKENHI Grant Numbers JP26810074 and JP16K15043.

REFERENCES (1)

Venne, A.S.; Kollipara, L.; Zahedi, R.P. The Next Level of Complexity: Crosstalk of Posttranslational Modifications. Proteomics 2014, 14, 513-524.

(2)

Pagel, O.; Loroch, S.; Sickmann, A.; Zahedi, R.P. Current Strategies and Findings in Clinically Relevant Post-Translational Modification-Specific Proteomics. Expert Rev. Proteomics 2015, 12, 235-253.

(3)

Silva, M.L. Cancer Serum Biomarkers Based on Aberrant Post-Translational Modifications of Glycoproteins: Clinical Value and Discovery Strategies. Biochim. Biophys. Acta. 2015, 1856, 165-177.

(4)

Walsh, G. Post-Translational Modifications of Protein Biopharmaceuticals. Drug Discov. Today 2010, 15, 773-780.

(5)

Walsh, G. Biopharmaceutical Benchmarks 2014. Nat. Biotechnol. 2014, 32, 992-1000.

(6)

Dumont, J.; Euwart, D.; Mei, B.; Estes, S.; Kshirsagar, R. Human Cell Lines for Biopharmaceutical Manufacturing: History, Status, and Future Perspectives. Crit. Rev. Biotechnol. 2016, 36, 1110-1122.

(7)

Rothbart, S.B.; Dickson, B.M.; Raab, J.R.; Grzybowski, A.T.; Krajewski, K.; Guo, A.H.; Shanle, E.K.; Josefowicz, S.Z.; Fuchs, S.M.; Allis, C.D.; Magnuson, T.R.; Ruthenburg, A.J.; Strahl, B.D. An Interactive Database for the Assessment of Histone Antibody Specificity. Mol. Cell 2015, 59, 502-511.

(8)

Hattori, T.; Lai, D.; Dementieva, I.S.; Montano, S.P.; Kurosawa, K.; Zheng, Y.; Akin, L.R.; Swist-Rosowska, K.M.; Grzybowski, A.T.; Koide, A.; Krajewski, K.; Strahl, B.D.; Kelleher, N.L.; Ruthenburg, A.J.; Koide, S. Antigen Clasping by Two Antigen-Binding Sites of An Exceptionally Specific Antibody for Histone Methylation. Proc. Natl. Acad. Sci. U. S. A. 2016, 113, 2092-2097.

(9)

Lin, W.; Gao, L.; Chen, X. Protein-Specific Imaging of Posttranslational Modifications. Curr. Opin. Chem. Biol. 2015, 28, 156-163.

(10) Fuchs, S.M.; Strahl, B.D. Antibody Recognition of Histone Post-Translational Modifications: Emerging Issues and Future Prospects. Epigenomics 2011, 3, 247-249.


19


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 20 of 32

(11) Hattori, T.; Taft, J.M.; Swist, K.M.; Luo, H.; Witt, H.; Slattery, M.; Koide, A.; Ruthenburg, A.J.; Krajewski, K.; Strahl, B.D.; White, K.P.; Farnham, P.J.; Zhao, Y.; Koide, S. Recombinant Antibodies to Histone Post-Translational Modifications. Nat. Methods 2013, 10, 992-995. (12) Martin, A.R.; Vasseur, J.J.; Smietana, M. Boron and Nucleic Acid Chemistries: Merging the Best of Both Worlds. Chem. Soc. Rev. 2013, 42, 5684-5713. (13) Kubota, R.; Hamachi, I. Protein Recognition Using Synthetic Small-Molecular Binders toward Optical Protein Sensing in vitro and in Live Cells. Chem. Soc. Rev. 2015, 44, 44544471. (14) Beaver, J.E.; Waters, M.L. Molecular Recognition of Lys and Arg Methylation. ACS Chem. Biol. 2016, 11, 643-653. (15) Minami, T.; Esipenko, N.A.; Akdeniz, A.; Zhang, B.; Isaacs, L.; Anzenbacher, P. Jr Multianalyte Sensing of Addictive Over-the-Counter (OTC) Drugs. J. Am. Chem. Soc. 2013, 135, 15238-15243. (16) Caglayan, M.G.; Sheykhi, S.; Mosca, L.; Anzenbacher, P. Fluorescent Zinc and Copper Complexes for Detection of Adrafinil in Paper-Based Microfluidic Devices. Chem. Commun. 2016, 52, 8279-8282. (17) Zhang, X.T.; Wang, S.; Xing, G.W. Boronlectin/Polyelectrolyte Ensembles as Artificial Tongue: Design, Construction, and Application for Discriminative Sensing of Complex Glycoconjugates from Panax ginseng. ACS Appl. Mater. Interfaces 2017, 9, 3368-3375. (18) Han, J.; Wang, B.; Bender, M.; Kushida, S.; Seehafer, K.; Bunz, U.H. Poly(aryleneethynylene) Tongue That Identifies Nonsteroidal Anti-Inflammatory Drugs in Water: A Test Case for Combating Counterfeit Drugs. ACS Appl. Mater. Interfaces 2017, 9, 790-797. (19) Miranda, O.R.; You, C.C.; Phillips, R.; Kim, I.B.; Ghosh, P.S.; Bunz, U.H.; Rotello, V.M. Array-Based Sensing of Proteins Using Conjugated Polymers. J. Am. Chem. Soc. 2007, 129, 9856-9857. (20) Li, H.; Bazan, G.C. Conjugated Oligoelectrolyte/ssDNA Aggregates: Self-Assembled Multicomponent Chromophores for Protein Discrimination. Adv. Mater. 2009, 21, 964967. (21) Wu, P.; Miao, L.N.; Wang, H.F.; Shao, X.G.; Yan, X.P. A Multidimensional Sensing Device for the Discrimination of Proteins Based on Manganese-Doped ZnS Quantum Dots. Angew. Chem. Int. Ed. 2011, 50, 8118-8121.


20

Page 21 of 32

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


(22) Pei, H.; Li, J.; Lv, M.; Wang, J.; Gao, J.; Lu, J.; Li, Y.; Huang, Q.; Hu, J.; Fan, C. A Graphene-Based Sensor Array for High-Precision and Adaptive Target Identification with Ensemble Aptamers. J. Am. Chem. Soc. 2012, 134, 13843-13849. (23) Motiei, L.; Pode, Z.; Koganitsky, A.; Margulies, D. Targeted Protein Surface Sensors as a Tool for Analyzing Small Populations of Proteins in Biological Mixtures. Angew. Chem. Int. Ed. 2014, 53, 9289-9293. (24) Selvakumar, K.; Motiei, L.; Margulies, D. Enzyme-Artificial Enzyme Interactions as a Means for Discriminating among Structurally Similar Isozymes. J. Am. Chem. Soc. 2015, 137, 4892-4895. (25) Wang, M.; Ye, H.; You, L.; Chen, X. A Supramolecular Sensor Array Using LanthanideDoped Nanoparticles for Sensitive Detection of Glyphosate and Proteins. ACS Appl. Mater. Interfaces. 2016, 8, 574-581. (26) Bajaj, A.; Miranda, O.R.; Kim, I.B.; Phillips, R.L.; Jerry, D.J.; Bunz, U.H.; Rotello, V.M. Detection and Differentiation of Normal, Cancerous, and Metastatic Cells Using Nanoparticle-Polymer Sensor Arrays. Proc. Natl. Acad. Sci. U. S. A. 2009, 106, 1091210916. (27) Bicker, K.L.; Sun, J.; Harrell, M.; Zhang, Y.; Pena, M.M.; Thompson, P.R.; Lavigne, J.J. Synthetic Lectin Arrays for the Detection and Discrimination of Cancer Associated Glycans and Cell Lines. Chem. Sci. 2012, 3, 1147-1156. (28) Rana, S.; Le, N.D.; Mout, R.; Duncan, B.; Elci, S.G.; Saha, K.; Rotello, V.M. A Multichannel Biosensor for Rapid Determination of Cell Surface Glycomic Signatures. ACS Cent. Sci. 2015, 1, 191-197. (29) Rana, S.; Elci, S.G.; Mout, R.; Singla, A.K.; Yazdani, M.; Bender, M.; Bajaj, A.; Saha, K.; Bunz, U.H.; Jirik, F.R.; Rotello, V.M. Ratiometric Array of Conjugated PolymersFluorescent Protein Provides a Robust Mammalian Cell Sensor. J. Am. Chem. Soc. 2016, 138, 4522-4529. (30) Tomita, S.; Yoshimoto, K. Polyion Complex Libraries Possessing Naturally Occurring Differentiation for Pattern-Based Protein Discrimination. Chem. Commun. 2013, 49, 10430-10432. (31) Tomita, S.; Soejima, T.; Shiraki, K.; Yoshimoto, K. Enzymatic Fingerprinting of Structurally Similar Homologous Proteins Using Polyion Complex Library Constructed by Tuning PEGylated Polyamine Functionalities. Analyst 2014, 139, 6100-6103. (32) Tomita, S.; Niwa, O.; Kurita, R. Artificial Modification of an Enzyme for Construction of Cross-Reactive Polyion Complexes to Fingerprint Signatures of Proteins and Mammalian Cells. Anal. Chem. 2016, 88, 9079-9086.


21


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 22 of 32

(33) Tomita, S.; Yokoyama, S.; Kurita, R.; Niwa, O.; Yoshimoto, K. The Use of an EnzymeBased Sensor Array to Fingerprint Proteomic Signatures of Sera from Different Mammalian Species. Anal. Sci. 2016, 32, 237-240. (34) Tomita, S.; Sakao, M.; Kurita, R.; Niwa, O.; Yoshimoto, K. A Polyion Complex Sensor Array for Markerless and Noninvasive Identification of Differentiated Mesenchymal Stem Cells from Human Adipose Tissue. Chem. Sci. 2015, 6, 5831-5836. (35) Zhang, T.; Edwards, N.Y.; Bonizzoni, M.; Anslyn, E.V. The Use of Differential Receptors to Pattern Peptide Phosphorylation. J. Am. Chem. Soc. 2009, 131, 11976-11984. (36) Minaker, S.A.; Daze, K.D.; Ma, M.C.; Hof, F. Antibody-Free Reading of the Histone Code Using a Simple Chemical Sensor Array. J. Am. Chem. Soc. 2012, 134, 11674-11680. (37) Zamora-Olivares, D.; Kaoud, T.S.; Dalby, K.N.; Anslyn, E.V. In-situ Generation of Differential Sensors that Fingerprint Kinases and the Cellular Response to Their Expression. J. Am. Chem. Soc. 2013, 135, 14814-14820. (38) Minami, T.; Esipenko, N.A.; Zhang, B.; Isaacs, L.; Anzenbacher, P. Jr "Turn-on" fluorescent sensor array for basic amino acids in water. Chem. Commun. 2014, 50, 61-63. (39) Mosca, L.; Karimi Behzad, S.; Anzenbacher, P. Jr Small-Molecule Turn-On Fluorescent Probes for RDX. J. Am. Chem. Soc. 2015, 137, 7967-7969. (40) Aoki, S.; Kawatani, H.; Goto, T.; Kimura, E.; Shiro, M. A Double-Functionalized Cyclen with Carbamoyl and Dansyl Groups (Cyclen = 1,4,7,10-Tetraazacyclododecane): A Selective Fluorescent Probe for Y3+ and La3+. J. Am. Chem. Soc. 2001, 123, 1123-1132. (41) Uchiyama, S.; Matsumura, Y.; de Silva, A.P.; Iwai, K. Fluorescent Molecular Thermometers Based on Polymers Showing Temperature-Induced Phase Transitions and Labeled with Polarity-Responsive Benzofurazans. Anal. Chem. 2003, 75, 5926-5935. (42) Hayashida, O.; Ogawa, N.; Uchiyama, M. Surface Recognition and Fluorescence Sensing of Histone by Dansyl-Appended Cyclophane-Based Resorcinarene Trimer. J. Am. Chem. Soc. 2007, 129, 13698-13705. (43) Suzuki, Y.; Takagi, N.; Sano, T.; Chimuro, T. Design and Synthesis of a Novel Fluorescent Protein Probe for Easy and Rapid Electrophoretic Gel Staining by Using a Commonly Available UV-Based Fluorescent Imaging System. Electrophoresis 2013, 34, 2464-2472. (44) Zhuang, Y.D.; Chiang, P.Y.; Wang, C.W.; Tan, K.T. Environment-Sensitive Fluorescent Turn-on Probes Targeting Hydrophobic Ligand-Binding Domains for Selective Protein Detection. Angew. Chem. Int. Ed. 2013, 52, 8124-8128.


22

Page 23 of 32

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


(45) Lienqueo, M.E.; Mahn, A.; Asenjo, J.A. Mathematical Correlations for Predicting Protein Retention Times in Hydrophobic Interaction Chromatography. J. Chromatogr. A 2002, 978, 71-79. (46) Majorek, K.A.; Porebski, P.J.; Dayal, A.; Zimmerman, M.D.; Jablonska, K.; Stewart, A.J.; Chruszcz, M.; Minor, W. Structural and Immunologic Characterization of Bovine, Horse, and Rabbit Serum Albumins. Mol. Immunol. 2012, 52, 174-182. (47) Liu, Y.; Bonizzoni, M. A Supramolecular Sensing Array for Qualitative and Quantitative Analysis of Organophosphates in Water. J. Am. Chem. Soc. 2014, 136, 14223-14229. (48) Anzenbacher, P. Jr; Lubal, P.; Bucek, P.; Palacios, M.A.; Kozelkova, M.E. A Practical Approach to Optical Cross-Reactive Sensor Arrays. Chem. Soc. Rev. 2010, 39, 3954-3979. (49) Tabachnick, B.G.; Fidell, L.S. Using Multivariate Statistics; Allyn & Bacon, Needham Heights, MA, 4th edn, 2001. (50) Kumpf, J.; Freudenberg, J.; Bunz, U.H. Distyrylbenzene-Aldehydes: Identification of Proteins in Water. Analyst 2015, 140, 3136-3142. (51) Lin, M.; Li, W.; Wang, Y.; Yang, X.; Wang, K.; Wang, Q.; Wang, P.; Chang, Y.; Tan, Y. Discrimination of Hemoglobins with Subtle Differences Using an Aptamer Based Sensing Array. Chem. Commun. 2015, 51, 8304-8306. (52) Schalkwijk, C.G.; Miyata, T. Early- and Advanced Non-enzymatic Glycation in Diabetic Vascular Complications: The Search for Therapeutics. Amino Acids 2012, 42, 1193-1204. (53) Anguizola, J.; Matsuda, R.; Barnaby, O.S.; Hoy, K.S.; Wa, C.; DeBolt, E.; Koke, M.; Hage, D.S. Review: Glycation of Human Serum Albumin. Clin. Chim. Acta. 2013, 425, 64-76. (54) Cohen, M.P. Diabetes and protein glycosylation. Measurement and biologic relevance; Springer Verlag, New York. 1986. (55) Derouiche, A.; Cousin, C.; Mijakovic, I. Protein Phosphorylation from the Perspective of Systems Biology. Curr. Opin. Biotechnol. 2012, 23, 585-590. (56) Roskoski, R. Jr ERK1/2 MAP Kinases: Structure, Function, and Regulation. Pharmacol. Res. 2012, 66, 105-143.


23


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 24 of 32

Figure for Table of Contents


24

Page 25 of 32

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. (A) Chemical structures of cationic poly-L-lysines (PLLs) modified with environment-sensitive fluorophores. (B) Schematic representation of fluorogenic interactions with a protein with/without PTMs. (C) Resulting fluorescence fingerprints reflecting each protein obtained under different solution conditions. 164x265mm (300 x 300 DPI)



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. Profiles of analyte proteins used in this study: (A) mammalian albumins; (B) bovine and human serum albumins with various chemical modifications corresponding to PTMs. For examples of these PTMs in organisms, see Table S1; (C) inactivated and activated GST-tagged human ERK1; Φsurface values of proteins were calculated based on the Miyazawa–Jernigan hydrophobicity scale.45 99x130mm (300 x 300 DPI)


Page 26 of 32

Page 27 of 32

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. Characterization of PLL-Dnc: (A) fluorescence spectra of 5.0 µg/mL PLL-Dnc upon addition of ESA in 18 mM MES (pH 5.5, λex = 340 nm, [ESA] = 0–59 µg/mL); (B-D) binding isotherms for 5.0 µg/mL PLLDnc upon addition of four mammalian serum albumins in (B) 18 mM MES (pH 5.5), (C) 18 mM MES (pH 5.5) + 25 mM NaCl, and (D) 18 mM EPPS (pH 8.5); λex = 340 nm, λem = 510 nm. 48x15mm (300 x 300 DPI)



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 4. Heat plot of the fluorescence fingerprints of 20 µg/mL albumins with/without chemical modifications obtained from the PLL-Dnc system. Six replicates are shown for each analyte. 53x15mm (300 x 300 DPI)


Page 28 of 32

Page 29 of 32

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 5. Discriminant score plot for 20 µg/mL albumins with/without chemical modifications corresponding to PTMs obtained from the PLL-Dnc system. 75x56mm (300 x 300 DPI)



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 6. Clustering analysis of discriminative elements of the PLL-Dnc system. The hierarchical clustering dendrogram based on the Euclidean distance with the Ward method was created using a data set (24 solutions × 13 analytes × 6 replicates) divided by the root mean square of the corresponding solution data. Areas "A" and "B" represent clustering of the same solution properties and same pH values, respectively. "C" indicates the distance between the clustering point in the presence/absence of NaCl at pH = 5.5 and that at pH = 7.0/8.5. 82x93mm (300 x 300 DPI)


Page 30 of 32

Page 31 of 32

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 7. Discriminant score plot for the quantitative response of (A) HSA and HSA-Glc with a total concentration of 0–30 µg/mL, (B) BSA and BSA-PS with a total concentration of 0–9 µg/mL, (C) ERK1-Ac and ERK1-In with a total concentration of 0–0.6 µg/mL using the PLL-Dnc system. The 95% confidence ellipses for the individual analytes are shown. 73x27mm (300 x 300 DPI)



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 for TOC 89x47mm (300 x 300 DPI)


Page 32 of 32

Fingerprinting Protein Populations with Post ... - ACS Publications

Recommend Documents