Synthesis and Applications of Boronate Affinity Materials: From Class

Aug 29, 2017 - Biography. Zhen Liu obtained his Ph.D. from Dalian Institute of Chemical Physics, Chinese Academy of Sciences, in 1998. He was appointe...
1 downloads 15 Views 3MB Size
This is an open access article published under an ACS AuthorChoice License, which permits copying and redistribution of the article or any adaptations for non-commercial purposes.

Article pubs.acs.org/accounts

Synthesis and Applications of Boronate Affinity Materials: From Class Selectivity to Biomimetic Specificity Zhen Liu* and Hui He State Key Laboratory of Analytical Chemistry for Life Science, School of Chemistry and Chemical Engineering, Nanjing University, Nanjing 210023, China CONSPECTUS: Due to the complexity of biological systems and samples, specific capture and targeting of certain biomolecules is critical in much biological research and many applications. cis-Diol-containing biomolecules, a large family of important compounds including glycoproteins, saccharides, nucleosides, nucleotides, and so on, play essential roles in biological systems. As boronic acids can reversibly bind with cisdiols, boronate affinity materials (BAMs) have gained increasing attention in recent years. However, real-world applications of BAMs are often severely hampered by three bottleneck issues, including nonbiocompatible binding pH, weak affinity, and difficulty in selectivity manipulation. Therefore, solutions to these issues and knowledge about the factors that influence the binding properties are of significant importance. These issues have been well solved by our group in past years. Our solutions started from the synthesis and screening of boronic acid ligands with chemical moieties favorable for binding at neutral and acidic pH. To avoid tedious synthesis routes, we proposed a straightforward strategy called teamed boronate affinity, which permitted facile preparation of BAMs with strong binding at neutral pH. To enhance the affinity, we confirmed that multivalent binding could significantly enhance the affinity toward glycoproteins. More interestingly, we observed that molecular interactions could be significantly enhanced by confinement within nanoscale spaces. To improve the selectivity, we investigated interactions that govern the selectivity and their interplays. We then proposed a set of strategies for selectivity manipulation, which proved to be useful guidelines for not only the design of new BAMs but also the selection of binding conditions. Applications in metabolomic analysis, glycoproteomic analysis, and aptamer selection well demonstrated the great potential of the prepared BAMs. Molecular imprinting is an important methodology for creating affinity materials with antibody-like binding properties. Boronate affinity-based covalent imprinting is a pioneering approach in molecular imprinting, but only a few cases of successful imprinting of glycoproteins by this method were reported. With sound understanding of boronate affinity, we developed two facile and generally applicable boronate affinity-based molecular imprinting approaches. The resulting boronate affinity molecularly imprinted polymers (MIPs) exhibited dramatically improved binding properties, including biocompatible binding pH range, enhanced affinity, improved specificity, and superb tolerance to interference. In terms of nanoconfinement effect, we explained why the binding pH range was widened and why the affinity was enhanced. The excellent binding properties made boronate affinity MIPs appealing alternatives to antibodies in promising applications such as disease diagnosis, cancer-cell targeting, and single-cell analysis. In this Account, we survey the key aspects of BAMs, the efforts we made to solve these issues, and the connections between imprinted and nonimprinted BAMs. Through this survey, we wish to pave a sound fundamental basis of the dependence of binding properties of BAMs on the nature and structure of the ligands and the supporting materials, which can facilitate the development and applications of BAMs. We also briefly sketch remaining challenges and directions for future development.



INTRODUCTION Much biological fundamental and applied research encounters a common bottleneck issue that compounds of interest exist in low abundance while a huge number of interfering compounds coexist in high abundance in the biological systems or samples. Therefore, specific recognition and capture of target compounds is often a key step prior to further studies. To this end, tremendous antibodies have been prepared. However, antibodies often suffer from limited sources, high price, and poor storage stability. Affinity materials have been widely used for the isolation of biological compounds. Among them, boronate affinity materials (BAMs) are unique, because the boronic acid ligands can reversibly bind with cis-diol-containing compounds, © 2017 American Chemical Society

such as glycoproteins, glycans, saccharides, nucleosides, and nucleotides. These cis-diol-containing biomolecules are important target molecules in current scientific frontiers such as proteomics, metabolomics, glycomics, glycobiology, and important applications such as disease diagnosis. BAMs have gained rapid development in the recent decade.1 Particularly, we have prepared a large variety of novel BAMs and demonstrated their great application prospects in important areas such as affinity separation, x-omic analysis, disease diagnosis, cancer-cell targeting, and single-cell analysis. Received: April 12, 2017 Published: August 29, 2017 2185

DOI: 10.1021/acs.accounts.7b00179 Acc. Chem. Res. 2017, 50, 2185−2193

Article

Accounts of Chemical Research

Affinity is another critical property of BAMs. Like other affinity materials, the binding constant of a ligand determines the target concentration at which the materials can be applicable. The dissociation constants (Kd) between boronic acids and sugars or glycoproteins range from 10−1 to 10−3 M.2 Since the concentration of cis-diol-containing biomolecules is usually very low, capture of them by conventional BAMs is rather difficult and even impossible. Selectivity is also an essential aspect. It was often challenging to obtain good selectivity particularly for glycoproteins with BAMs. A major reason for this is that multiple interactions governing the selectivity, and their interplays were not taken into account in the design and application of BAMs. To achieve good selectivity, a sound understanding of the interaction mechanism is indispensable.

This Account surveys the key issues that BAMs encounter, the efforts and strategies we used to solve these issues, and new methodologies we developed to prepare biomimetic materials with desirable binding properties, particularly boronate affinitybased molecular imprinting. More importantly, we try to unveil the dependence of binding properties of BAMs on the nature and structure of the ligands and the supporting materials.



BORONATE AFFINITY AND KEY ASPECTS Boronate affinity relies on reversible covalent interaction between boronic acids and cis-diol-containing compounds, as shown in Figure 1. Generally, when the surrounding pH is



NOVEL BORONATE AFFINITY MATERIALS To solve the three key issues, in the past decade, we prepared a series of new BAMs with improved binding properties, which permitted promising applications. Binding pH

Four types of boronic acids can provide low binding pH, including (I) phenylboronic acids with electron-withdrawing groups, such as sulfonyl, fluoro, and carbonyl, on the phenyl ring,3,4 (II) boronic acids containing intramolecular tetracoordinated B−N or B−O bonds (Wulff-type),5,6 (III) boronic acids containing intramolecular tricoordinated B−O bonds (improved Wulff-type),7,8 and (IV) heterocyclic boronic acids.9 BAMs functionalized with these four types of boronic acids and their binding pH are shown in Figure 2.

Figure 1. Interaction between boronic acids and cis-diol-containing compounds.

equal to or greater than the pKa value of the boronic acid, the boronic acid can adduct with hydroxyl and form a tetragonal boronate anion (sp3) and thereby can react with cis-diols and form five- or six-membered cyclic esters. When the surrounding pH is greatly lower than the pKa value of the boronic acid, the boronic acid−cis-diol complex dissociates because the boronic acid completely reverts to trigonal configuration (sp2). Such chemistry endows boronic acid-functionalized materials with several important merits, including relatively stronger affinity (as compared with nonspecific interactions), broad-spectrum selectivity (a ligand can bind a large variety of compounds), pH-controlled capture and release, and fast association/ desorption kinetics. Particularly, because a mild acidic solution can well desorb target molecules from the material, boronate affinity-based separation and sample pretreatment is quite compatible with mass spectrometry (MS)-based x-omic analysis, in which acidic conditions are often used for ionization. However, conventional BAMs are associated with three apparent drawbacks, including (1) nonbiocompatible binding pH, (2) weak affinity, and (3) relatively poor selectivity, which largely hamper their real-world applications. Binding pH is a key factor for BAMs. Although the total number of commercial boronic acids and derivatives is more than 1500, only a limited number are suitable for the synthesis of BAMs. 3-Aminophenylboronic acid (APBA, pKa = 8.8) and 4-vinylphenylboronic acid (VPBA, pKa = 8.2), which contain an active moiety, are the most widely used ones. Thus, conventional BAMs can only function well under alkaline conditions. Since the pH of frequently used biosamples, such as blood, tear, urine, and saliva, ranges from 4.5 to 8.0, the application to these samples requires adjusting the sample to a basic pH, which gives rise to not only the inconvenience of pH adjustment but also the risk of degradation of labile biomolecules.

Figure 2. BAMs with biocompatible binding pH.

Boronic acids with electron-withdrawing groups possess increased acidity and lowered pKa value. We first synthesized 4(3-butenylsulfonyl) phenylboronic acid (BSPBA; pKa = 7.0) and prepared a BSPBA-functionalized monolithic capillary.10 The capillary allowed for binding nucleosides and glycoproteins at neutral pH. In addition, we prepared a silica hybrid monolith functionalized with 3-acrylamidophenylboronic acid (AAPBA).3 It exhibited a binding pH of 6.5. Moreover, 2,4-difluoro-3formyl-phenylboronic acid (DFFPBA) was screened from commercial boronic acids as a ligand candidate with low 2186

DOI: 10.1021/acs.accounts.7b00179 Acc. Chem. Res. 2017, 50, 2185−2193

Article

Accounts of Chemical Research

Figure 3. Teamed boronate affinity formed through (A) ring-opening polymerization and (B) molecular self-assembly. Adapted with the permission from ref 1. Copyright 2015 Royal Society of Chemistry.

(TBA).14,15 As shown in Figure 3, TBA turned on a “molecular team” that comprises a regular boronic acid (such as APBA and thiophene-3-boronic acid) and a neighboring diamine (such as 1,6-hexamethylenediamine) or linear difunctional amine (such as 2-mercaptoethylamine), rather than a single boronic acid. By virtue of boron−nitrogen coordination between the boronic acid and the amine-containing compound, the two compounds could form a molecular team. Through an appropriate reaction such as ring-opening polymerization14 or molecular selfassembly,15 the molecular team was fixed in a supporting material, and the resulting TBA material could function as a single Wulff-type boronic-acid ligand that could bind cis-diolcontaining compounds at neutral pH. Although the binding pH of TBA materials is not so low as compared with most of the above-mentioned BAMs, the TBA strategy provides a facile and straightforward access to prepare BAMs that can function at neutral pH. Using this strategy, we prepared a protein Amimicking material, called restricted access boronate affinity porous monolith, for the specific isolation of immunoglobulin G (IgG).16

binding pH.2 We prepared three DFFPBA-functionalized monolithic columns with varying spacer arms. 11 The monolithic column with appropriate spacer arm could bind cis-diol-containing compounds at pH 6.0. Wulff-type boronic acids contain intramolecular tetracoordinated B−N or B−O bonds, which facilitate the formation of tetragonal boronate anion (sp3). The sp3 hybridization status remains stable even under neutral or moderately acidic conditions, facilitating boronate esterification with cis-diols. Wulff-type boronic acids can provide a pKa value as low as 5.2. However, most Wulff-type boronic acids reported in literature did not carry a reactive moiety for immobilization. We synthesized 3-(dimethylaminomethyl) aniline-4-pinacol boronate.12 The boronate was attached to a monolithic column. After rinsing with an acidic solution, the pinacol moiety of the boronate was removed, and the resulting monolithic column exhibited a binding pH as low as 5.5. Benzoboroxoles, improved Wulff-type boronic acids, show excellent water solubility and improved sugar binding capability.5,6 We synthesized 3-carboxybenzoboroxole and immobilized it onto a monolithic column.13 Although the pKa value was not significantly low (6.9), the 3-carboxybenzoboroxole-immobilized monolithic column exhibited a binding pH as low as 5.0. This suggests that benzoboroxoles may not follow the general binding pH−acidity relationship. Heterocyclic boronic acids exhibit apparently low binding pH due to the presence of a heteroatom in the ring. 3Pyridinylboronic acid exhibits a pKa value of 4.4. We prepared a 3-pyridinylboronic acid-functionalized monolithic column.9 It permitted selective binding of cis-diol-containing compounds at pH as low as 4.5, the lowest binding pH so far among BAMs. Such a binding pH enabled the enrichment of cis-diol containing biomolecules such as nucleosides from urine samples without pH adjustment. If not commercially available, boronic acids with favorable structures must be synthesized in-lab, which usually requires tedious synthesis procedures. For instance, the synthesis of 3(dimethylaminomethyl) aniline-4-pinacol boronate required four reaction steps.12 To avoid tedious synthesis routes, we proposed a novel strategy called teamed boronate affinity

Affinity

For a certain boronic acid ligand, the affinity depends on their structure as well as the pH used. Even at the most favorable pH, the affinity of boronic acids toward cis-diol-containing biomolecules is still low. Therefore, affinity enhancement strategies are of significant importance. Biomolecules such as antibodies can strongly bind with their target molecules, because of their avidity through synergistic multiple binding rather than the affinity of a single binding. Using dendrimeric boronic acid-functionalized magnetic nanoparticles (MNPs), we confirmed the effectiveness of the multivalent binding strategy in enhancing the binding toward glycoproteins.17 As shown in Figure 4, due to the presence of poly(amidoamine) dendrimer (generation 4.0, 64 surface amino groups), the number of boronic acid moieties at the surface of the MNPs was greatly amplified. For compounds containing only one single cis-diol, such as monosaccharides and nucleosides, the Kd values with the MNPs were nearly the same as those for monovalent binding, at the 10−2 M level. 2187

DOI: 10.1021/acs.accounts.7b00179 Acc. Chem. Res. 2017, 50, 2185−2193

Article

Accounts of Chemical Research

Figure 4. Enhanced affinity through multivalent binding. Adapted with the permission from ref 17. Copyright 2013 Royal Society of Chemistry.

Figure 5. Molecular interactions under different spatially confined conditions. Adapted with the permission from ref 21. Copyright 2014 Royal Society of Chemistry.

However, for glycoproteins such as horseradish peroxidase (HRP) and transferrin (TRF), due to the synergistic binding between multiple boronic acids on the MNPs and multiple sugars of the glycoproteins, the Kd values were enhanced by 3− 4 orders of magnitude as compared with the single binding affinity toward monosaccharides, reaching the 10−5 to 10−6 M level. To further improve the avidity, we used a dendrimer with flexible branched chains, polyethylenimine (PEI), as a bindingsite-amplifying scaffold to prepare boronate avidity monolithic capillary.18 The prepared capillary exhibited Kd values of 10−6 to 10−7 M toward glycoproteins, allowing for the selective enrichment of trace glycoproteins in human saliva. Apparently, the multivalent binding strategy does not work for compounds containing only a single cis-diol. Therefore, another effective strategy is of importance for improving the binding affinity toward this type of compound. When evaluating the performance of affinity materials, in addition to the properties of the ligands, the effects of the material structures on the performance must be taken into account. When a compound is confined in a limited space, its physicochemical properties, such as density19 and catalytic activity,20 may dramatically differ from those under nonconfined conditions. However, knowledge about the nanoconfinement effect on molecular interactions in affinity materials had not been established before. To this end, we carried out a quantitative study on molecular interactions confined within boronate affinity mesoporous silica nanoparticles (Figure 5).21 We observed that the molecular interactions were enhanced by the confinement and the enhancement factor depended on the confinement degree. Under a high confinement (a 1.5 nm sized glycoprotein, RNase B, confined in 2.1 nm boronate affinity nanopores), the affinity was enhanced by 3 orders of magnitude as compared with nonconfined conditions. While under less confinement (adenosine, a small molecule containing only one cis-diol, confined in 2.6 nm pores), the boronate affinity was enhanced by only 7.5-fold. Such a finding implies that when a target molecule is confined in a space that is complementary to the target in 3D shape, which is a typical situation in molecularly imprinted polymers (MIPs), the binding strength will be enhanced to the maximum by the nanoconfinement effect. The two aspects discussed above, that is, multivalent binding and confinement effect, work independently. Therefore, a more

efficient strategy is to combine the two aspects. Such a strategy is useful for not only molecular imprinting but also the design of new affinity materials. According to this strategy, we designed and prepared a novel type of affinity materials called nanoconfining affinity materials.22 These materials relied on the confinement effect of porous materials to provide key affinity for proteins with molecular sizes comparable to the pore sizes, enabling promising applications including depletion of highabundance serum proteins, chiral separation, and immobilized enzyme reactors. Selectivity

Liu23 first surveyed the molecular interactions involved in boronate affinity chromatography. From our first study on the design and synthesis of BAMs, we investigated the selectivitycontrolling molecular interactions and their interplays.24 Since then, knowledge obtained from previous studies were further used for subsequent design and preparation of new BAMs.25 With increased knowledge and understanding, we proposed a suite of strategies for selectivity adjustment.26 As illustrated in Figure 6, besides boronate affinity interaction, multiple secondary interactions, primarily including hydrophobic interaction, ionic interaction, and hydrogen bonding, can occur on BAMs, while Lewis base effect (Lewis acid−base coordination) favors complexation under less basic conditions. Therefore, the strategies to obtain good selectivity involve two aspects. On one hand, strong boronate affinity interaction is crucial, for which boronic acid ligands with higher affinity and lower binding pH are essential. Meanwhile, the involvement of Lewis base can enhance the affinity and thereby improve the selectivity. On the other hand, unfavorable secondary interactions should be suppressed, for which useful means are suggested in Figure 6. These strategies have been verified as useful guidelines for not only the design of new BAMs10,24,25 but also the selection of binding conditions.12−14 If being exquisitely manipulated, secondary interactions can act as favorable factors. Such a possibility has enabled promising secondary separations after boronate affinity separation. Using hydrogen bonding interaction as a major driving force for secondary separation, highly hydrophilic boronate affinity monolithic columns have been designed, which enabled twodimension (2D) separations of cis-diol compounds in a single column.3,10 The first dimension separation was boronate 2188

DOI: 10.1021/acs.accounts.7b00179 Acc. Chem. Res. 2017, 50, 2185−2193

Article

Accounts of Chemical Research

Figure 6. Strategies for selectivity manipulation. Green arrows, favorable interaction; cyan arrows, unfavorable interactions; red up arrow, the interaction can be enhanced by the means specified; blue down arrows, the interactions can be suppressed by the means specified. Adapted with the permission from ref 26. Copyright 2015 Wiley.

affinity separation, in which cis-diol-containing compounds were retarded on the columns. When the retarded compounds were eluted with an acidic solution, the components in the retarded fraction were further separated according to their hydrogen bonding interaction with the column. In addition, a 3-pyridylboronic acid-functionalized organic−silica hybrid monolithic column enabled secondary anion-exchange separation of negatively charged cis-diol-containing compounds such as ribonucleotides.9

Another representative application is aptamer selection. Aptamers are single-stranded DNA or RNA oligonucleotides capable of specifically binding certain targets, making them attractive alternatives to antibodies. Nucleic acid aptamers are usually selected from a large random oligonucleic acid library by systematic evolution of ligands by exponential enrichment (SELEX).28,29 However, conventional SELEX methods are associated with apparent disadvantages including tedious procedure (typically 8−16 rounds, 2−4 weeks), large reagent consumption, and nonspecific binding. Using the 3-carboxybenzoboroxole-functionalized monolithic capillary,13 we developed a boronate affinity capillary-based SELEX approach.30 It permitted selection of glycoprotein-binding aptamers by 6 rounds within 2 days. This approach provided several merits, including speed, specificity, and minute reagent consumption. Recently, we further developed a hybrid aptamer selection approach that combined boronate affinity MNPs and capillary electrophoresis (CE).31 It allowed for selection of glycoproteinbinding aptamers in 4 rounds within 2 days. The selected aptamers exhibited excellent affinity and specificity and enabled enzyme activity assay of alkaline phosphatase (ALP) in clinical serum samples.

Applications

A typical application area of BAMs is glycoproteomics. The 3carboxybenzoboroxole-functionalized monolithic column13 can well demonstrate the class-selectivity and efficiency of BAMs for protein glycosylation analysis. After selective enrichment using this column, the obtained compounds were sent to matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) MS for identification. From the tryptic digests of HRP, a total of 22 glycopeptides were effectively enriched and identified, as shown in Figure 7. As a comparison, if no enrichment was performed prior to the analysis, only 2 glycopeptides were identified.27



BORONATE AFFINITY MOLECULARLY IMPRINTED POLYMERS Although BAMs with greatly improved binding properties have shown great application prospects, they are still associated with two limitations. Due to their class selectivity, they fail to recognize specific cis-diol-containing compounds. In addition, most of the materials still suffer from relatively poor affinity (Kd, 10−5−10−7 M). To this end, molecular imprinting is naturally an effective solution, as indicated by above discussion. Molecular Imprinting

Molecular imprinting is a practical methodology to synthesize artificial receptors to mimic enzymes and antibodies at the material level,32−35 which has enabled a number of important applications such as catalysis, sensing, and separation. The imprinting methodologies can be classified into two types: covalent imprinting pioneered by Wulff32 and noncovalent imprinting pioneered by Mosbach.33 Noncovalent imprinting is more flexible regarding the selection of functional monomers and possible target molecules. However, it may yield certain

Figure 7. Comparison of mass spectra for the analysis of tryptic digests of HRP by MALDI-TOF MS without and with boronate affinity enrichment. Adapted with the data reported in ref 27. 2189

DOI: 10.1021/acs.accounts.7b00179 Acc. Chem. Res. 2017, 50, 2185−2193

Article

Accounts of Chemical Research

appropriate porogen (pH ≥ 8.0). The template and the monomer formed covalent complexes via boronate affinity, driving molecular self-assembly. The complexes were then mixed with the cross-linker and a UV initiator. Through UV radiation for a short duration (tens of seconds), the obtained prepolymer rapidly polymerized into a MIP. Such fast polymerization avoided apparent conformational variation of the template. The imprinted template in the MIP was facilely removed by extraction with an acidic solution, leaving behind imprinted cavities complementary to the molecular shape of the template. In this way, molecularly imprinted thin-layer arrays were easily prepared through a photolithographic procedure. The prepared boronate affinity MIP arrays exhibited multiple highly favorable features, including biocompatible binding pH range (5.0−9.0), strong affinity (Kd ca. 85 nM), high specificity (cross-reactivity ≤8.8%), and superb tolerance for interference (tolerating the existence of competing monosaccharide of one million-fold higher concentration).

heterogeneity of binding sites due to the existence of equilibrium in prepolymerization. In contrast, covalent imprinting generates a more even binding-site distribution owing to the stability of covalent interactions. In addition, the structures of guest-binding sites are more understandable in covalent imprinting. Because the easy on/off reactivity of boronic acids favors imprinting and removal of templates, boronic acids have been promising functional monomers for the covalent imprinting of cis-diol-containing compounds. Since boronate affinity-based molecular imprinting of small molecules was first reported by Wulff and co-workers,32 a number of MIPs against small molecules such as monosaccharides and nucleotides have been prepared.36,37 The imprinting of biomacromolecules, especially proteins, is more challenging. This is because the large size of proteins makes it difficult to remove them from highly cross-linked polymer networks. It is also because harsh polymerization conditions used for imprinting leads to conformational change of proteins. To overcome these issues, many imprinting strategies were proposed, including surface imprinting,38 epitope imprinting,39 Pickering emulsions,40 metal coordination,41 hierarchical imprinting, 42 and so on. The covalent imprinting of glycoproteins using boronic acids was first proposed by Mosbach and co-workers in 1985.43 They prepared transferrin-imprinted polysiloxane-coated silica. However, the imprinting of other glycoproteins with similar approach failed owing to problems such as protein precipitation. In 2013, Chen et al.44 and Li et al.45 independently reported boronate affinitybased surface imprinting of glycoproteins. However, the advantages of boronate affinity-based MIPs were not unveiled until our first study on this direction published.

Boronate Affinity Based Controllable Oriented Surface Imprinting

Although the above imprinting approach demonstrated highly desirable binding properties, it is still associated with a limitation; that is, the substrates must be able to accept UV radiation. To solve this issue, we further developed another facile and general but more applicable approach, boronate affinity based controllable oriented surface imprinting.46 The principle is illustrated in Figure 9. A glycoprotein template is first immobilized on a boronic acid-functionalized substrate surface via boronate affinity binding. Then, a hydrophilic imprinting layer is gradually formed by in-water self-polymerization of dopamine and APBA to an appropriate thickness (as a rule of thumb, 1/3 to 2/3 of the size of the template molecule in one dimension). After template removal, cavities complementary to the molecular shape of the template are created in the imprinting layer. This approach can be applied to diverse substrates. The in-water self-polymerization reaction allows for imprinting labile proteins directly in aqueous buffer solution averting their conformational change or denaturation. Because the thickness of the imprinting layer linearly depended on the polymerization time, it is easy to control the thickness of imprinting layer to obtain desirable properties through adjusting the imprinting time. The prepared MIPs inherited all the excellent binding properties of above boronate affinitybased MIPs. Moreover, because the imprinting was implemented on the surface, effective mass transferring facilitated template removal and target rebinding, high imprinting efficiency (48.5%) was obtained. Very interestingly, MIPs produced by this method could rebind the templates in dual modes: at a high pH that enabled boronate affinity interaction, the rebinding was due to the combination of shape matching and boronate affinity binding (high affinity mode, Kd ca. 10−9 M); while at an acidic pH that disrupted the boronate affinity interaction, the rebinding was mainly due to strong nanoconfinement (low affinity mode, Kd ca. 10−7 M). Such a mechanism explains well why boronate affinity MIPs can bind target molecules over a much wider pH range and provide much higher binding affinity as compared with nonimprinted counterparts. We further developed a protocol that allows for precise control of the thickness of the imprinting layer via adjusting the imprinting time, by which glycans could be facilely imprinted through calculating the imprinting time according to the

Photolithographic Boronate Affinity Molecular Imprinting

We developed a general and facile approach, photolithographic boronate affinity molecular imprinting, for preparing MIPs specific to glycoproteins.45 As illustrated in Figure 8, the principle relied on UV-initiated free radical polymerization of a cross-linker (e.g., poly(ethylene glycol) diacrylate) with a boronic acid monomer (e.g., VPBA). The template was first mixed with the monomer in an alkaline solution of an

Figure 8. Schematic of photolithographic boronate affinity molecular imprinting. Adapted with the permission from ref 45. Copyright 2013 Wiley. 2190

DOI: 10.1021/acs.accounts.7b00179 Acc. Chem. Res. 2017, 50, 2185−2193

Article

Accounts of Chemical Research

Figure 9. Schematic of boronate affinity-based controllable oriented surface imprinting. Reproduced with the permission from ref 46. Copyright 2014 Royal Society of Chemistry.

Figure 10. (A) An image showing in vivo extraction within a single living cell. (B) Raman intensity of ALP in individual cells of different cell lines and ALP solutions (1.5 U/L, 5 mL). Adapted with the permission from ref 55. Copyright 2016 Wiley.

targeting agents for cell or tissue imaging.49,50 Single-cell analysis techniques are essential for unveiling the microheterogeneity and behaviors of cells. Particularly, low-copy number proteins (present at fewer than 1000 molecules per cell) play important roles in cell functioning, such as signaling and the regulation of gene expression. However, the determination of low-copy-number proteins in single living cells remains challenging. Recently, a boronate affinity MIP was integrated onto an in vivo extraction microprobe; through combination with plasmon-enhanced Raman scattering (PERS), probing of low-copy-number ALP in single living cells was achieved (Figure 10).55

molecular length of the templates as well as the boronic acid ligand.47 The prepared MIPs could specifically recognize an intact glycoprotein and its characteristic fragments, even within a complex sample matrix, providing particularly valuable sorbents for the extraction of unstable glycoproteins. Glycans have been known to pose serious challenges to antibody development.48 Besides, glycans are usually not or weakly immunogenic. In this sense, boronate affinity controllable oriented surface imprinting opened a new avenue for efficient preparation of antibody mimics for the recognition of glycans and according glycoproteins. This was the first report that molecular imprinting was performed in a precisely controllable fashion. Inspired by this work, we further prepared monosaccharide-imprinted nanoscale MIPs using this approach. The prepared MIPs exhibited good specificity toward the target monosaccharides.49,50



CONCLUDING REMARKS Through this Account, we hope to shed light on fundamental aspects that determine the properties and application capability of BAMs. Such understanding can facilitate the development and applications of BAMs. From the viewpoint of binding properties, the imprinted BAMs reported recently are definitely more advantageous than nonimprinted ones. They benefit from three fundamental aspects: boronate affinity, multivalent binding, and nanoconfinement. However, from the viewpoint of application, the two categories of materials could be equally important, since they are complementary to each other; that is, nonimprinted materials provide class selectivity for binding a large groups of target compounds, while imprinted materials excel at targeting specific compounds. As to future development, nonimprinted BAMs with significantly improved affinity remain critical for wide practical applications, while the design and preparation of boronate affinity MIPs for targeting biologically significant glycoproteins and glycans such as biomarkers on cancer cells and receptors in disease signaling pathways can be a challenging and promising direction. In addition, the combination of BAMs with other advanced

Applications

Since boronate affinity MIPs exhibited highly favorable binding properties, a range of promising applications, which are impossible with nonimprinted BAMs, have been enabled. One application is disease diagnosis. Glycoprotein-imprinted microarrays were used as substitutes of antibodies for enzymelinked immunosorbent assay (ELISA) of trace α-fetoprotein (AFP) in human serum.45 Soon, through combining glycoprotein-imprinted microarrays and surface-enhanced Raman scattering (SERS), a novel antibody-free and enzyme-free approach, called boronate affinity sandwich assay (BASA), was constructed, and the determination of trace glycoprotein biomarkers in human serum was accomplished.51 In addition, glycoprotein-imprinted 96-well plates and substrates were also employed for specific extraction of trace glycoprotein biomarkers from serum and urine samples.52−54 Moreover, monosaccharide-imprinted fluorescent and Raman-active nanoparticles demonstrated particular strength as cancer-cell2191

DOI: 10.1021/acs.accounts.7b00179 Acc. Chem. Res. 2017, 50, 2185−2193

Article

Accounts of Chemical Research

(9) Li, D.; Li, Q.; Wang, S.; Ye, J.; Nie, H.; Liu, Z. Pyridinylboronic Acid-Functionalized Organic-Silica Hybrid Monolithic Capillary for the Selective Enrichment and Separation of cis-Diol-Containing Biomolecules at Acidic pH. J. Chromatogr. A 2014, 1339, 103−109. (10) Liu, Y.; Ren, L.; Liu, Z. A Unique Boronic Acid Functionalized Monolithic Capillary for Specific Capture, Separation and Immobilization of cis-Diol Biomolecules. Chem. Commun. 2011, 47, 5067−5069. (11) Li, Q.; Lu, C.; Liu, Z. Preparation and Characterization of Fluorophenylboronic Acid-Functionalized Monolithic Columns for High Affinity Capture of cis-Diol Containing Compounds. J. Chromatogr. A 2013, 1305, 123−130. (12) Li, H.; Liu, Y.; Liu, J.; Liu, Z. A Wulff-Type Boronate for Boronate Affinity Capture of cis-Diol Compounds at Medium Acidic pH Condition. Chem. Commun. 2011, 47, 8169−8171. (13) Li, H.; Wang, H.; Liu, Y.; Liu, Z. A BenzoboroxoleFunctionalized Monolithic Column for the Selective Enrichment and Separation of cis-Diol Containing Biomolecules. Chem. Commun. 2012, 48, 4115−4117. (14) Ren, L.; Liu, Z.; Liu, Y.; Dou, P.; Chen, H.-Y. Ring-Opening Polymerization with Synergistic Co-Monomers: Access to a BoronateFunctionalized Polymeric Monolith for the Specific Capture of cisDiol-Containing Biomolecules under Neutral Conditions. Angew. Chem. 2009, 121, 6832−6835. (15) Liang, L.; Liu, Z. A Self-Assembled Molecular Team of Boronic Acids at the Gold Surface for Specific Capture of cis-Diol Biomolecules at Neutral pH. Chem. Commun. 2011, 47, 2255−2257. (16) Liu, Y.; Lu, Y.; Liu, Z. Restricted Access Boronate Affinity Porous Monolith as a Protein A Mimetic for the Specific Capture of Immunoglobulin G. Chem. Sci. 2012, 3, 1467−1471. (17) Wang, H.; Bie, Z.; Lü, C.; Liu, Z. Magnetic Nanoparticles with Dendrimer-Assisted Boronate Avidity for the Selective Enrichment of Trace Glycoproteins. Chem. Sci. 2013, 4, 4298−4303. (18) Li, D.; Li, Y.; Li, X.; Bie, Z.; Pan, X.; Zhang, Q.; Liu, Z. A High Boronate Avidity Monolithic Capillary for the Selective Enrichment of Trace Glycoproteins. J. Chromatogr. A 2015, 1384, 88−96. (19) Heuberger, M.; Zach, M.; Spencer, N. D. Density Fluctuations under Confinement: When is a Fluid not a Fluid? Science 2001, 292, 905−908. (20) Fu, Q.; Li, W.-X.; Yao, Y.; Liu, H.; Su, H.-Y.; Ma, D.; Gu, X.-K.; Chen, L.; Wang, Z.; Zhang, H.; Wang, B.; Bao, X. Interface-Confined Ferrous Centers for Catalytic Oxidation. Science 2010, 328, 1141− 1144. (21) Chen, Y.; Wang, S.; Ye, J.; Li, D.; Liu, Z.; Wu, X. Insights into the Effect of Nanoconfinement on Molecular Interactions. Nanoscale 2014, 6, 9563−9567. (22) Li, Q.; Tu, X.; Ye, J.; Bie, Z.; Bi, X.; Liu, Z. Nanoconfining Affinity Materials for pH-Mediated Protein Capture-Release. Chem. Sci. 2014, 5, 4065−4069. (23) Liu, X.-C. Boronic Acids as Ligands for Affinity Chromatography. Chin. J. Chromatogr. 2006, 24, 73−80. (24) Ren, L.; Liu, Y.; Dong, M.; Liu, Z. Synthesis of Hydrophilic Boronate Affinity Monolithic Capillary for Specific Capture of Glycoproteins by Capillary Liquid Chromatography. J. Chromatogr. A 2009, 1216, 8421−8425. (25) Ren, L.; Liu, Z.; Dong, M.; Ye, M.; Zou, H. Synthesis and Characterization of a New Boronate Affinity Monolithic Capillary for Specific Capture of cis-Diol-Containing Compounds. J. Chromatogr. A 2009, 1216, 4768−4774. (26) Muhammad, P.; Li, D., Liu, Z. Boronate Affinity Chromatography. In Encyclopedia of Analytical Chemistry; Meyers, R. A., Ed; John Wiley & Sons, Ltd: California, USA, 2015. (27) Bie, Z.; Chen, Y.; Li, H.; Wu, R.; Liu, Z. Off-Line Hyphenation of Boronate Affinity Monolith-Based Extraction with Matrix-Assisted Laser Desorption/Ionization Time-of-Flight Mass Spectrometry for Efficient Analysis of Glycoproteins/Glycopeptides. Anal. Chim. Acta 2014, 834, 1−8. (28) Ellington, A. D.; Szostak, J. W. In-vitro Selection of RNA Molecules that Bind Specific Ligands. Nature 1990, 346, 818−822.

materials (such as plasmonic nanomaterials and quantum dots) and desirable response (such as photothermal and Raman scattering) will greatly expand the applications of BAMs.



AUTHOR INFORMATION

Corresponding Author

*Tel: +86-25-8968-5639. E-mail: [email protected]. ORCID

Zhen Liu: 0000-0002-8440-2554 Funding

This work was supported by the National Science Fund for Distinguished Young Scholars (No. 21425520) from the National Natural Science Foundation of China, the “333” Talents Project from Jiangsu Provincial Government (No. BRA2016351), and the Open Grant from the State Key Laboratory of Analytical Chemistry for Life Science (No. 5431ZZXM1605). Notes

The authors declare no competing financial interest. Biographies Zhen Liu obtained his Ph.D. from Dalian Institute of Chemical Physics, Chinese Academy of Sciences, in 1998. He was appointed as a full professor at Nanjing University in 2005. He was awarded the National Science Fund for Distinguished Young Scholars in 2014. His current major research interest is to develop advanced functional materials and innovative approaches for molecular recognition, separation, bioanalysis, and disease diagnosis. Hui He received his B.S. degree (2012) and M.S. degree (2015) from Northwestern A&F University, China. Currently, he is a Ph.D. candidate in Prof. Zhen Liu’s group at Nanjing University. His research focuses on coupling of boronate affinity extraction with mass spectrometry.



REFERENCES

(1) Li, D.; Chen, Y.; Liu, Z. Boronate Affinity Materials for Separation and Molecular Recognition: Structure, Properties and Applications. Chem. Soc. Rev. 2015, 44, 8097−8123. (2) Lü, C.; Li, H.; Wang, H.; Liu, Z. Probing the Interactions between Boronic Acids and cis-Diol-Containing Biomolecules by Affinity Capillary Electrophoresis. Anal. Chem. 2013, 85, 2361−2369. (3) Li, Q.; Lu, C.; Li, H.; Liu, Y.; Wang, H.; Wang, X.; Liu, Z. Preparation of Organic-Silica Hybrid Boronate Affinity Monolithic Column for the Specific Capture and Separation of cis-Diol Containing Compounds. J. Chromatogr. A 2012, 1256, 114−120. (4) Li, X.; Pennington, J.; Stobaugh, J. F.; Schöneich, C. Synthesis of Sulfonamide- and Sulfonyl-Phenylboronic Acid-Modified Silica Phases for Boronate Affinity Chromatography at Physiological pH. Anal. Biochem. 2008, 372, 227−236. (5) Wulff, G. Selective Binding to Polymers via Covalent Bonds. The Construction of Chiral Cavities as Specific Receptor Sites. Pure Appl. Chem. 1982, 54, 2093−2102. (6) Wulff, G.; Lauer, M.; Böhnke, H. Rapid Proton Transfer as Cause of an Unusually Large Neighboring Group Effect. Angew. Chem., Int. Ed. Engl. 1984, 23, 741−742. (7) Bérubé, M.; Dowlut, M.; Hall, D. G. Benzoboroxoles as Efficient Glycopyranoside-Binding Agents in Physiological Conditions: Structure and Selectivity of Complex Formation. J. Org. Chem. 2008, 73, 6471−6479. (8) Dowlut, M.; Hall, D. G. An Improved Class of Sugar-Binding Boronic Acids, Soluble and Capable of Complexing Glycosides in Neutral Water. J. Am. Chem. Soc. 2006, 128, 4226−4227. 2192

DOI: 10.1021/acs.accounts.7b00179 Acc. Chem. Res. 2017, 50, 2185−2193

Article

Accounts of Chemical Research (29) Tuerk, C.; Gold, L. Systematic Evolution of Ligands by Exponential Enrichment: RNA Ligands to Bacteriophage T4 DNA Polymerase. Science 1990, 249, 505−510. (30) Nie, H.; Chen, Y.; Lu, C.; Liu, Z. Efficient Selection of Glycoprotein-Binding DNA Aptamers via Boronate Affinity Monolithic Capillary. Anal. Chem. 2013, 85, 8277−8283. (31) Li, X.; He, Y.; Ma, Y.; Bie, Z.; Liu, B.; Liu, Z. A Hybrid Approach Combining Boronate Affinity Magnetic Nanoparticles and Capillary Electrophoresis for Efficient Selection of GlycoproteinBinding Aptamers. Anal. Chem. 2016, 88, 9805−9812. (32) Wulff, G.; Sarhan, A. The Use of Polymers with EnzymeAnalogous Structures for the Recognition of Racemates. Angew. Chem. 1972, 84, 364. (33) Vlatakis, G.; Andersson, L. I.; Müller, R.; Mosbach, K. Drug Assay using Antibody Mimics Made by Molecular Imprinting. Nature 1993, 361, 645−647. (34) Wulff, G. Molecular Imprinting in Cross-Linked Materials with the Aid of Molecular TemplatesA Way towards Artificial Antibodies. Angew. Chem., Int. Ed. Engl. 1995, 34, 1812−1832. (35) Shi, H.; Tsai, W. B.; Garrison, M. D.; Ferrari, S.; Ratner, B. D. Template-Imprinted Nanostructured Surfaces for Protein Recognition. Nature 1999, 398, 593−597. (36) Sallacan, N.; Zayats, M.; Bourenko, T.; Kharitonov, A. B.; Willner, I. Imprinting of Nucleotide and Monosaccharide Recognition Sites in Acrylamidephenylboronic Acid-Acrylamide Copolymer Membranes Associated with Electronic Transducers. Anal. Chem. 2002, 74, 702−712. (37) Gu, L.; Jiang, X.; Liang, Y.; Zhou, T.; Shi, G. Double Recognition of Dopamine Based on a Boronic Acid Functionalized Poly(Aniline-Co-Anthranilic Acid) Molecularly Imprinted Polymer Composite. Analyst 2013, 138, 5461−5469. (38) Kempe, M.; Glad, M.; Mosbach, K. An Approach towards Surface Imprinting using the Enzyme Ribonuclease A. J. Mol. Recognit. 1995, 8, 35−39. (39) Nishino, H.; Huang, C. S.; Shea, K. J. Selective Protein Capture by Epitope Imprinting. Angew. Chem., Int. Ed. 2006, 45, 2392−2396. (40) Shen, X.; Ye, L. Molecular Imprinting in Pickering Emulsions: A New Insight into Molecular Recognition in Water. Chem. Commun. 2011, 47, 10359−10361. (41) Qin, L.; He, X.; Zhang, W.; Li, W.; Zhang, Y. Macroporous Thermosensitive Imprinted Hydrogel for Recognition of Protein by Metal Coordinate Interaction. Anal. Chem. 2009, 81, 7206−7216. (42) Nematollahzadeh, A.; Sun, W.; Aureliano, C. S.; Lutkemeyer, D.; Stute, J.; Abdekhodaie, M. J.; Shojaei, A.; Sellergren, B. High-Capacity Hierarchically Imprinted Polymer Beads for Protein Recognition and Capture. Angew. Chem., Int. Ed. 2011, 50, 495−498. (43) Glad, M.; Norrlow, O.; Sellergren, B.; Siegbahn, N.; Mosbach, K. Use of Silane Monomers for Molecular Imprinting and Enzyme Entrapment in Polysiloxane-Coated Porous Silica. J. Chromatogr. A 1985, 347, 11−23. (44) Li, Y.; Hong, M.; Miaomiao; Bin, Q.; Lin, Z.; Cai, Z.; Chen, G. Novel Composites of Multifunctional Fe3O4@Au Nanofibers for Highly Efficient Glycoprotein Imprinting. J. Mater. Chem. B 2013, 1, 1044−1051. (45) Li, L.; Lu, Y.; Bie, Z.; Chen, H.-Y.; Liu, Z. Photolithographic Boronate Affinity Molecular Imprinting: A General and Facile Approach for Glycoprotein Imprinting. Angew. Chem., Int. Ed. 2013, 52, 7451−7454. (46) Wang, S.; Ye, J.; Bie, Z.; Liu, Z. Affinity-Tunable Specific Recognition of Glycoproteins via Boronate Affinity-Based Controllable Oriented Surface Imprinting. Chem. Sci. 2014, 5, 1135−1141. (47) Bie, Z.; Chen, Y.; Ye, J.; Wang, S.; Liu, Z. Boronate-Affinity Glycan-Oriented Surface Imprinting: A New Strategy to Mimic Lectins for the Recognition of an Intact Glycoprotein and Its Characteristic Fragments. Angew. Chem., Int. Ed. 2015, 54, 10211− 10215. (48) Sterner, E.; Flanagan, N.; Gildersleeve, J. C. Perspectives on Anti-Glycan Antibodies Gleaned from Development of a Community Resource Database. ACS Chem. Biol. 2016, 11, 1773−1783.

(49) Yin, D.; Wang, S.; He, Y.; Liu, J.; Zhou, M.; Ouyang, J.; Liu, B.; Chen, H.-Y.; Liu, Z. Surface-Enhanced Raman Scattering Imaging of Cancer Cells and Tissues via Sialic Acid-Imprinted Nanotags. Chem. Commun. 2015, 51, 17696−17699. (50) Wang, S.; Yin, D.; Wang, W.; Shen, X.; Zhu, J. J.; Chen, H. Y.; Liu, Z. Targeting and Imaging of Cancer Cells via MonosaccharideImprinted Fluorescent Nanoparticles. Sci. Rep. 2016, 6, 22757. (51) Ye, J.; Chen, Y.; Liu, Z. A Boronate Affinity Sandwich Assay: An Appealing Alternative to Immunoassays for the Determination of Glycoproteins. Angew. Chem., Int. Ed. 2014, 53, 10386−10389. (52) Bi, X.; Liu, Z. Facile Preparation of Glycoprotein-Imprinted 96Well Microplates for Enzyme-Linked Immunosorbent Assay by Boronate Affinity-Based Oriented Surface Imprinting. Anal. Chem. 2014, 86, 959−966. (53) Bi, X.; Liu, Z. Enzyme Activity Assay of Glycoprotein Enzymes Based on a Boronate Affinity Molecularly Imprinted 96-Well Microplate. Anal. Chem. 2014, 86, 12382−12389. (54) Tu, X.; Muhammad, P.; Liu, J.; Ma, Y.; Wang, S.; Yin, D.; Liu, Z. Molecularly Imprinted Polymer-Based Plasmonic Immunosandwich Assay for Fast and Ultrasensitive Determination of Trace Glycoproteins in Complex Samples. Anal. Chem. 2016, 88, 12363−12370. (55) Liu, J.; Yin, D.; Wang, S.; Chen, H.-Y.; Liu, Z. Probing LowCopy-Number Proteins in a Single Living Cell. Angew. Chem., Int. Ed. 2016, 55, 13215−13218.

2193

DOI: 10.1021/acs.accounts.7b00179 Acc. Chem. Res. 2017, 50, 2185−2193