Preparation and Characterization of Biofunctionalized Inorganic

Jul 2, 2015 - Lauren J. Webb is an associate professor of chemistry at The University of Texas at Austin. She obtained her A.B. in chemistry (music mi...
3 downloads 8 Views 7MB Size
Invited Feature Article pubs.acs.org/Langmuir

Preparation and Characterization of Biofunctionalized Inorganic Substrates Jason W. Dugger and Lauren J. Webb* Department of Chemistry, Center for Nano- and Molecular Science and Technology, and Institute for Cell and Molecular Biology, The University of Texas at Austin, 105 E. 24th Street, STOP A5300, Austin, Texas 78712-1224, United States ABSTRACT: Integrating the function of biological molecules into traditional inorganic materials and substrates couples biologically relevant function to synthetic devices and generates new materials and capabilities by combining biological and inorganic functions. At this so-called “bio/abio interface,” basic biological functions such as ligand binding and catalysis can be co-opted to detect analytes with exceptional sensitivity or to generate useful molecules with chiral specificity under entirely benign reaction conditions. Proteins function in dynamic, complex, and crowded environments (the living cell) and are therefore appropriate for integrating into multistep, multiscale, multimaterial devices such as integrated circuits and heterogeneous catalysts. However, the goal of reproducing the highly specific activities of biomolecules in the perturbed chemical and electrostatic environment at an inorganic interface while maintaining their native conformations is challenging to achieve. Moreover, characterizing protein structure and function at a surface is often difficult, particularly if one wishes to compare the activity of the protein to that of the dilute, aqueous solution phase. Our laboratory has developed a general strategy to address this challenge by taking advantage of the structural and chemical properties of alkanethiol self-assembled monolayers (SAMs) on gold surfaces that are functionalized with covalently tethered peptides. These surface-bound peptides then act as the chemical recognition element for a target protein, generating a biomimetic surface in which protein orientation, structure, density, and function are controlled and variable. Herein we discuss current research and future directions related to generating a chemically tunable biofunctionalization strategy that has potential to successfully incorporate the highly specialized functions of proteins onto inorganic substrates.



INTRODUCTION Biological processes occur through complex interactions that depend on chemical identity, covalent and noncovalent interactions, molecular structure, and environmental conditions, codependent factors that have arisen from a long series of evolutionary developments. The inherent extraordinary specificity of biological molecules for a given function, such as ligand binding or catalysis, makes them attractive systems to exploit for a variety of applications including molecular electronics, sensing, and commercial green catalysis integrated into biologically relevant situations.1−4 The range of functions of biomolecules is largely irreproducible by artificial devices or technologies. Because of this, attention has turned to strategies that attempt to incorporate, rather than replicate, biological functions in an artificial or “abiological” context. While the functions of biomolecules are exquisitely refined to interact with other molecules and their environment, this specificity becomes an obstacle to successfully integrating biologically based materials with inorganic substrates while maintaining native activity. As biological function is inherently dependent on the structure and dynamics of the biomolecule, it is necessary to design devices that are able to preserve native biological conformations while retaining the unique properties of the inorganic support so that these biological processes may be exploited in some useful manner. A textbook example of this is the electrochemical glucose biosensor, which employs the enzyme © XXXX American Chemical Society

glucose oxidase to oxidize glucose in human blood and generate hydrogen peroxide. This product is detected electrochemically as a calibrated read-out to determine the amount of glucose in a complex sample of whole blood.5 The speed, sensitivity, and selectivity of this device have revolutionized the home monitoring and care of diabetes to the profound benefit of patients. The use of such devices has become widespread, with applications ranging from the food industry to biomedical use, all based on the function of an enzyme immobilized at a surface.6,7 Such success warrants the research of a general immobilization technique that can be applied to a broad class of proteins for a variety of applications. A number of creative approaches have been demonstrated to enable this idea, including physical adsorption, covalent coupling, affinity tagging, cross-linking, and the use of biocompatible polymers.8 (While this review focuses exclusively on coupling proteins to surfaces, readers interested in the emerging field of coupling DNA structures to surfaces are directed toward other reviews.9−11) While these methodologies have successfully demonstrated the immobilization of functional biomolecules in specific cases, our laboratory has been guided by the goal of developing immobilization protocols that simultaneously satisfy the conditions of controlled immobilization, structural Received: May 21, 2015 Revised: June 29, 2015

A

DOI: 10.1021/acs.langmuir.5b01876 Langmuir XXXX, XXX, XXX−XXX

Langmuir

Invited Feature Article

Figure 1. (A) Cartoon model depicting a biofunctionalized surface that lacks control over protein orientation and contains a thick blocking layer that defines the chemical properties of the surface and reduces the underlying substrate to merely a passive support. (B) Model surface that employs a relatively thin, chemically tunable SAM to covalently tether small peptides to the substrate while preserving conformation and orientation. (C) Protein surface attachment through native protein−protein interactions to control protein orientation, structure, and function.

substrate. Self-assembled monolayers (SAMs) are formed on gold surfaces to produce a chemical supporting layer that can be easily modified with a wide variety of chemical functionality. Some fraction of this monolayer carries a terminating azide group which reacts with peptides containing alkyne or nitrile groups through a Huisgen cycloaddition (“click”) reaction that covalently binds the peptides to the SAM through a tri- or tetra-azole linkage.16,17 This protocol allows unprecedented control over the chemistry of the surface as the conformation and amino acid identity of the peptides can be selectively defined on the basis of sequence and amino acid side chain identity (natural or unnatural). Our group has demonstrated the synthesis and characterization of substrates functionalized with peptides in either α-helical or β-stranded conformations.18−21 The chemistry and characterization of these surfaces is the subject of this review. The work thus far represents a necessary prerequisite for the goal of peptide-mediated biofunctionalization through the exploitation of the native structural and electrostatic factors that dominate protein−protein interactions in the living cell.

preservation, and imposed orientation to generate well-defined, reproducible surfaces and that are generalizable to many systems of interest performing a wide variety of applications in both biological and artificial environments, shown schematically in Figure 1. Adsorption or entrapment using biocompatible polymer layers can be nonspecific and does not always allow for controlled orientation of the biomolecule, which is an important factor in preserving function at an interface (Figure 1A). Additionally, these biocompatible blocking layers can be significantly larger in dimension than the protein itself, resulting in a surface defined by the chemistry and physical properties of the blocking layer, not the substrate or protein. Covalent coupling results in nonreversible protein attachment while affinity binding, such as biotin/avidin or hexa-histidine, requires specific functional groups to be present on the biomolecules.12−14 An inherent limitation in all of these techniques is the lack of generalizability to a wide variety of proteins; a new synthetic protocol must be established for every protein under consideration, with no guarantee of success. In consideration of these limitations, an alternative path to biofunctionalization that may prove more successful is one that builds upon the previous strategies mentioned here but also exploits factors that dictate protein activity in vivo. Because the structures of proteins and other biomolecules are sensitive to their local environment, it is challenging to preserve their natural, solution-phase functionality at the interface of an inorganic substrate. Protein conformations and interactions often rely solely on noncovalent forces dictated by strong, localized electrostatic fields that vary in magnitude and direction over relatively small distances on the surface of the protein.15 One way to overcome this obstacle is to chemically modify the substrate in such a way as to mimic natural biomolecular interactions found in the living cell. Considering the native method of protein−protein interactions, which rely on structural and electrostatic complementarity, one option for immobilizing functional biomolecules would be to employ the use of small peptides bound covalently to the inorganic substrate composed of amino acid sequences that would interact with and specifically noncovalently bind a protein introduced from solution (Figure 1B,C). This method of using peptides as the chemical recognition element of a biofunctionalization scheme would have the inherent advantage of producing surfaces that have electrostatic complementarity to target proteins at length scales relevant to their native functional forms in a selective, oriented, and recyclable manner (Figure 1C). With this strategy in mind, our group has developed a protocol for covalently tethering short peptides to inorganic surfaces while maintaining complete control over the chemical properties of the



CHEMISTRY The immobilization strategy reviewed here uses gold-coated inorganic substrates that are subsequently functionalized with a SAM that contains terminal azide groups that react with alkyne or nitrile groups present on peptides to form a tri- or tetra-azole linkage (Figure 2) through a Huisgen cycloaddition (“click” chemistry). In optimizing this methodology for the goals described above, our laboratory has focused on creating a generalizable platform that allows extensive control over the chemical properties and structure of all aspects of the system. As designed, the protocol can be altered to use a variety of inorganic substrates, SAM lengths, chemical functionalities, and peptides containing both natural and unnatural amino acids. By establishing a biofunctionalization method that allows broad, tunable control over all aspects of the system, the design of the platform allows protein immobilization through native complementary chemical recognition. Substrate. We selected gold for our initial studies because it offers a conductive surface that is easily functionalized with the alkanethiol SAM and can be used on a variety of substrates including mica, quartz, and silicon wafers, depending on the particular application or characterization technique of interest. For instance, the macroscopic flatness of silicon wafers makes them useful for infrared spectroscopy experiments in reflection geometry, the atomic flatness of mica is ideal for scanning probe microscopy, and the UV transparency of quartz makes it suitable for measurements in the UV−visible region of the spectrum such as surface circular dichroic (CD) spectroscopy. Similar B

DOI: 10.1021/acs.langmuir.5b01876 Langmuir XXXX, XXX, XXX−XXX

Langmuir

Invited Feature Article

Figure 2. Schematic outlining the preparation of surfaces terminated with α-helical or β-stranded peptides beginning with 25% Br-terminated SAM, which allows for ∼20 Å spacing between terminal Br groups. The Br terminal groups are subsequently replaced with N3, which then undergo a cycloaddition reaction with alkyne groups (highlighted in blue) present on the peptide to form triazole linkages that tether the peptide to the SAM. As the distance between alkyne groups on the helical peptide correspond to the spacing of reactive N3 groups on the surface, the helical conformation of the peptide is induced through reaction with the supporting substrate while the β-stranded peptide continues to self-assemble with other β-strands to form fibrillar structures.

helical peptides, a 25%:75% solution of AzUDT/DT was used to space surface reactive azide groups approximately 20 Å apart. The click reaction is carried out in a 2:1 tert-butanol/H2O solvent, where sodium ascorbate is used to reduce Cu(II) to Cu(I) in situ, which then acts as the catalyst in the reaction between the azide groups on the surface and the alkyne groups on the peptide to form a triazole linkage. As the reaction takes place in an aqueous environment, TBTA (tris[(1-benzyl-1H1,2,3-triazol- 4-yl)methyl]amine) is used to stabilize the Cu(I) ions to prevent them from oxidizing back to Cu(II) in solution.26 Because SAMs are relatively stable at neutral pH, the solution chemistry of the click reaction does not greatly affect the stability or ordering of the SAM, although any possible damage to the SAM is checked at all points along the sample preparation through infrared and X-ray photoelectron spectroscopies (described below). A successful alternative approach to these reactions that we have explored used peptides containing nitrile groups which reacted with the azide-terminated SAM to form a tetrazole linkage in the presence of diethyl azodicarboxylate. As the presence of copper in these reactions could potentially interfere with electronic applications or create a cytotoxic environment, copper-free click reactions using cyclooctynes could be explored to address these issues.27 This versatility in the choice of the method of covalent tethering is an important aspect of the generalizability of this platform. Research in our laboratory is currently focused on examining the influence that charged terminal groups on the SAM have on the binding and orientation of helices reacted with the substrate. We have demonstrated that using SAMs composed of 25% AzUDT and 75% DT results in the successful attachment of structured α-helices that are free of aggregates or other unwanted debris. Studies in progress in our laboratory investigate surfaces of 25% AzUDT and 75% X compositions, where X represents

considerations can be taken into account for the function of the resulting surface, which might require atomic flatness or substrate electrical conductivity. Additionally, the generation of patterned surfaces through photolithography or microcontact insertion printing could be used to selectively functionalize areas of a surface with different peptides, possibly creating devices with multiple or complementary catalytic functions.22 Other well-established surface chemistries such as siloxanes on SiO2 substrates are in principle completely compatible with our general methodology and will be explored in future research. Self-Assembled Monolayers. The alkane-thiol-based SAM introduces a relatively thin, well-ordered layer where the terminal functional groups can be chosen both to mediate covalent binding to peptides and to define the chemical properties of the bulk surface. The thiol groups of a SAM react with gold atoms of the surface in an RS-Au-SR manner, yielding a (3 × 4) phase which transforms to (3 × 2(3)1/2)-rect. because of noncovalent van der Waals interactions between the longer alkyl chain tails of the SAM. For longer alkyl chains, such as those used in these studies, the alkyl chain flexibility can allow the terminal SAM groups to organize into (31/2 × 31/2)R30°, resulting in uniform coverage with a 5 Å distance between adjacent terminal groups.23,24 Monolayers composed of two or more different thiols are simple to create by using mixtures of thiols in the original deposition solution. For example, we commonly use a controlled ratio of 11-azido-1-undecanethiols (AzUDTs) and decanethiols (DTs) to create a surface terminated in azide groups where the distance between adjacent AzUDTs can be estimated by the ratio of component thiols on the surface. Although not all bicomponent thiol mixtures result in mixed SAMs without phase segregation on the surface, minimizing the difference between the alkyl chain lengths of the components aids in the formation of fully mixed SAMs.25 For reactions with C

DOI: 10.1021/acs.langmuir.5b01876 Langmuir XXXX, XXX, XXX−XXX

Langmuir

Invited Feature Article

environment for the peptide. What is not known, however, is if the SAM along with the covalent linkages is solely responsible for the relatively long structural stability in ambient air or if there are other stabilizing factors such as the presence of water molecules near the peptides. For fibrillar structures in particular, it is known that water does play a role in the rate of fibrillation, but it is unclear how it impacts the stability of the final structures.29,30 Research in this area is underway in our laboratory. Collectively, the degree of chemical flexibility inherent in this protocol allows for substantial control over all aspects of the system. The choice of the supporting inorganic substrate can be used to cater to specific characterization techniques, the composition of terminal groups of the SAM can be tailored to dictate where covalent reactions occur, and the sequence of the peptide accommodates not only the secondary structure at the surface but also complete flexibility for which residues are exposed to the aqueous interface. These factors all work in concert to establish the foundation of a tunable biofunctionalization strategy that preserves structure, imposes orientation, and results in a well-defined surface of immobilized biomolecules.

either hydrophobic, hydrophilic, or positively and negatively charged surfaces. These results will guide future experimental designs by categorizing the effect that the bulk terminal chemistry of the SAM at the interface has on the binding orientation, repulsion, and attraction experienced by helical peptides. In applications where the solution in which the sample is immersed is composed of several different biomolecules, optimization of the terminal group composition of the SAM could be used to prevent nonspecific interactions with unwanted biomolecules, thus further promoting the specific binding of a particular target protein. With a mind toward application, our laboratory is also investigating one-pot synthesis methods for the simultaneous generation of the SAM and tethered peptide to achieve peptide functionalization in a faster and more direct manner. Surface Attachment of Peptide. After the SAM has been formed, the next step is the covalent binding of short α-helical and β-stranded peptides to the substrate. Considering the ultimate goal of generating surfaces functionalized with active proteins, preserving the conformation of surface-bound peptides is integral to ensuring structural and electrostatic complementarity to the target protein surface. To demonstrate this capability, we have focused our efforts on peptides built from alternating sequences of leucine and lysine residues to generate either α-helical (LKKLXKKLLKKLLKKXLKKL) or β-stranded (KLKXKLLLKXKLK) structures, where the hydrophobic periodicities were chosen to encourage folding of the polypeptide into each conformation.28 The amino acids denoted by X are unnatural propargylglycine residues containing the reactive alkyne groups required for the click reaction. The goal is not to take a perfectly structured helix in solution and simply add it to the surface but instead to create surface-bound conditions that generate a stable helix structure in the peptide. In a stable α-helix, the distance and relative orientation of each side chain is well known, as hydrogen bonding between N−H and CO groups occurs between i and i + 4 residues, yielding a helical pitch of 5.4 Å. The propargylglycine residue can therefore be placed at the appropriate position in the peptide sequence to locate the two reactive alkyne groups on the same side of the helix and the same distance apart as the azides on the surface of the monolayer (Figure 2). This allows peptide secondary structure to be engineered into the surface chemistry directly, without relying on solution-phase structures which may or may not be significantly perturbed in the presence of the surface. Peptide design for β-sheets is simpler because of their straight geometry, so reaction conditions such as the azide surface density, time, temperature, and identity of the cycloaddition catalyst can be modified to generate fibrillar structures composed of aggregated β-sheets, where the N−H group of one β-strand is hydrogen bonded to a CO of an adjacent strand and side chains of each residue alternate in an up/down configuration along each strand. Taken together, we have demonstrated that both individual peptides and larger, macromolecular structures that self-assemble in vivo can also be immobilized at inorganic interfaces through chemisorption in a straightforward and easily reproducible manner. While the chemical assembly of the system takes place in solution, the substrates are dried and stored in ambient air until they can be characterized. One observation that has come from the development of this protocol is that the structure and attachment of the peptides are stable in ambient air for weeks after preparation. This is a promising result as we have established that the structure of a biomolecule is significantly dependent on its local environment, indicating that the SAM of the system seems to help in satisfying the creation of a favorable



CHARACTERIZATION In any device that attempts to co-opt the function of a biological molecule, the structure and activity of that molecule must be fully characterized and understood. This is often challenging because standard solution-phase methodologies for characterizing protein structure and function are often difficult to translate to surfaces and interfacial structures. This idea pertains not just to the biomolecule but also to the complex substrate on which it is based and whose surface chemistry is vital for controlling the biomolecule itself. The characterization of SAMs has been thoroughly explored in a number of diverse fields, so its use here as the supporting layer for biomolecular immobilization allows a wide range of techniques to be applied to study any chemical modifications made to the substrates with excellent surface sensitivity. Our strategy has been to employ a variety of surfacespecific techniques including grazing incidence angle reflection− absorption infrared spectroscopy (GRAS-IR), X-ray photoelectron spectroscopy (XPS), ellipsometry, scanning tunneling microscopy (STM), and atomic force microscopy (AFM) to gain a complete structural and chemical picture of the system. No single experimental methodology allows complete characterization of the surface, but taken together, a comprehensive picture may be accumulated that includes both surfaceaveraged and molecular-level information. Vibrational spectroscopy in the reflection−absorption geometry provides information about the chemical composition and structure of surface-associated functional groups, with the methyl and amide regions often containing the most useful information about substrate composition and orientation. The methyl and methylene peaks of the IR spectrum are useful in determining not only SAM formation and ordering but also surface attachment of the peptides seen through correlated changes in peak intensity. Depending on the combinations of thiols used to form the SAM, substrates may contain symmetric and asymmetric methylene bands (near 2850 and 2918 cm−1) arising from the alkyl chains as well as symmetric and asymmetric methyl bands (near 2879 and 2966 cm−1) if −CH3 is chosen as the terminal group of the thiol. Figure 3 shows representative spectra of a 100% AzUDTterminated surface (blue) with methylene peaks at 2851 and 2924 cm−1, along with a control sample that lacked catalyst (red) and a β-peptide reacted sample (black); all spectra were collected with an InSb detector. The symmetry of each peak indicates D

DOI: 10.1021/acs.langmuir.5b01876 Langmuir XXXX, XXX, XXX−XXX

Langmuir

Invited Feature Article

Figure 3. GRAS-IR spectra of the methyl region of a 100% AzUDT SAM (blue), a 100% AzUDT control sample which lacked catalyst (red), and a β-peptide-reacted surface (black). All samples contain symmetric and asymmetric methylene stretches near 2851 and 2924 cm−1, respectively, while the peptide-reacted sample also contains methyl bands at 2870 and 2958 cm−1. Spectra are offset for clarity. Adapted from ref 18. Copyright 2015 American Chemical Society.

Figure 4. Spectra from the amide region of GRAS-IR scans taken from SAMs reacted with the helical peptide (red) as well as the β-stranded peptide (black). The amide I peak shows strong absorbance at 1661 cm−1, indicating helical secondary structure (red), while splitting of the peak into two bands at 1695 and 1630 cm−1 reflects the antiparallel β-sheet conformation of the β-peptide at the surface (black). Spectra are offset for clarity. Adapted from refs 18 and 20. Copyright 2015 and 2012 American Chemical Society.

ordering within the alkyl chains of the SAM and can be monitored in control samples to ensure that the click reaction does not destroy or greatly perturb the SAM. The peptide-terminated surface shows peak growth at 2870 and 2958 cm−1, corresponding to the leucine methyl groups introduced by the bonding of the peptide to the surface. Along with the increase in intensity of the methylene bands and the reduced peak symmetry of the spectrum, the functional groups introduced by the peptides are not as well ordered as the alkyl groups in the underlying SAM, as expected by our model in Figure 2. The amide region of the IR spectrum (1400−1800 cm−1) is especially useful for not only confirming the presence of peptides on the surface but also providing insight into their secondary structures.31 Reaction success is confirmed by the presence of the amide I (∼1660 cm−1) and amide II (∼1550 cm−1) bands, where spectra were collected on an MCT detector and shown in Figure 4. The amide I peak arises predominantly from stretching of the backbone carbonyl groups, and to a lesser extent, C−N stretches and C−C−N deformations, while amide II is due to inplane bending of N−H, C−N stretching, in-plane bending of CO, as well as C−C and N−C stretching.32 The amide I band is especially sensitive to the secondary structure of peptides because of the way common configurations align carbonyl groups in particular orientations to maximize intrapeptide (α-helix) and interpeptide (β-sheet) hydrogen bonding that define stable secondary structures. The carbonyl groups within helical structures are all roughly pointed in the same direction and hydrogen bonded to N−H groups four residues away along the backbone. Transition dipole coupling of the amide groups produces a strong vibrational band near 1661 cm−1 for surfaces terminated with a bonded α-helix (Figure 4, red). As the carbonyl groups of helices are oriented in known directions, it is possible to calculate the angle of the helical axis relative to the surface through eq 1 1 1 2⎡⎣ 2 (3 cos2 θ − 1)]⎤⎦⎡⎣ 2 (3 cos2 θI − 1)⎤⎦ + 1 D=K 1 1 2⎡⎣ 2 (3 cos2 θ − 1)⎤⎦⎡⎣ 2 (3 cos2 θII − 1)⎤⎦ + 1

where D is the ratio between the integrated areas of the amide I and II bands on the SAM, K is a proportionality constant calculated from the ratio of integrated areas of amide II to amide I from a solution of the randomly oriented peptide, θI is the angle between the amide I mode and the helix backbone, θII is the angle between the amide II mode and the backbone, and θ is the angle between the surface normal and the helical axis.33,34 Our optimized peptide attachment chemistry described above results in surfaces with a helical axis of 37 ± 11° with respect to the surface (Figure 4, red). We interpret this result as approximately two-thirds of the surface-bound peptides connected by two triazoles as shown in Figure 2, while approximately one-third of the surface contains peptides that are bound by only one triazole linkage. For surfaces reacted with β-stranded peptides, the amide I peak can shift to a lower energy or split depending on whether the peptides at the surface aggregate into parallel or antiparallel β-sheets as those structures have different geometries of their backbone carbonyl groups. For the β-stranded peptides used in these studies, the amide I peak splits into two bands at 1695 and 1630 cm−1, corresponding to the B1 and B2 modes, respectively. The presence of these modes is characteristic of coupled transition dipole moments of carbonyl groups that indicate the formation of antiparallel β-sheets (Figure 4, black). While a complete discussion of the origin of vibrational bands in peptides and proteins is beyond the scope of this article, we direct the reader to another review for more thorough analysis.31 XPS has proven to be a successful tool for monitoring the peptide reaction as well as the composition of the SAM. XPS is a quantitative technique that measures the elemental composition of the sample surface and can be used to differentiate between the types of bonds those elements participate in. The C 1s region of the XP spectrum can be used to confirm SAM formation, where signal from only aliphatic C−C bonds is detected at 284.5 eV. Subsequent reaction of the surface with a peptide results in the growth of peaks corresponding to C−(N,O) and CO bonds at 285.8 and 287.7 eV, respectively, seen clearly in both

(1) E

DOI: 10.1021/acs.langmuir.5b01876 Langmuir XXXX, XXX, XXX−XXX

Langmuir

Invited Feature Article

Figure 6. CD spectra of the helical (red) and β-stranded (black) peptides in solution (dashed lines) and bound to the surface (solid lines). While the helical peptide only has a weak degree of helicity in solution, upon binding with the surface a more rigid helix is formed, as indicated by the troughs at 209 and 223 nm. Similarly, the β-peptide adopts a random-coil configuration in solution but takes on a β-sheet secondary structure when bound to the surface, as characterized by the trough at 218 nm. Adapted from refs 18 and 20. Copyright 2015 and 2012 American Chemical Society.

Figure 5. XP spectra of the C 1s binding region of SAMs reacted with the helical (red) and β-stranded (black) peptides. As the peptides share similar elemental compositions, the two spectra are almost identical, containing aliphatic C−C (284.5 eV), C−(N,O) (285.8 eV), and CO (287.7 eV) bonds. Spectra are offset for clarity. Adapted from refs 18 and 20. Copyright 2015 and 2012 American Chemical Society.

α- and β-peptide-terminated surfaces in Figure 5. Additionally, the azide groups of the SAM have distinctly different binding energies compared to those of N−H groups on the peptides that are observed in the N 1s region, thus that signal could be used in the future to quantify the degree of surface coverage of bound peptides.35 Solution and surface CD measurements have also been used to determine how the conformation of the peptides changes when bound to the surface. Because peptides are chiral molecules that absorb in the UV region, their CD spectra can be used to identify whether they adopt random-coil, helical, or β-sheet configurations. In solution, CD is commonly used to quickly assess the extent of secondary structure in a soluble protein, but we have adapted this capability to characterize surface-bound biomolecules as well.36,37 Quartz is used as the supporting substrate in these experiments because of its UV transparency, and three substrates are stacked and immersed in solution within a typical quartz cuvette to maximize the signal arising from the chiral absorption of the immobilized biomolecules. This technique then allows for a direct comparison of secondary structures between the solution phase and surface bound states of a biomolecule. For example, the peptide LKKLXKKLLKKLLKKXLKKL was found to be only weakly helical in room-temperature aqueous solutions. However, when bound to the surface, this peptide folded into an induced α-helical structure, shown by intense negative peaks at 209 and 223 nm in the CD spectrum (Figure 6, red). This was particularly remarkable because the peptide click reaction took place under unfolding conditions of 80 °C for 5 h. This result indicates that the surface plays a role in inducing a helical conformation of the peptide when it is bound to the surface, the result of the structural design parameters of the system, where a combination of defined spacing of terminal azide groups on the SAM as well as the spacing of the reactive alkyne groups within the peptide give rise to a desired secondary structure at the surface. It is worth reiterating that such results are a consequence of the tunable control over all components in the system. Similar observations have been made for the peptide KLKXKLLLKXKLK. The alternating sequence of K and L

creates a polar peptide that should form a stable strandlike structure, although CD spectra clearly show that this peptide adopts a random-coil configuration in aqueous solution (negative band