SEPTEMBER/OCTOBER 1996 Volume 7, Number 5 © Copyright 1996 by the American Chemical Society
COMMUNICATIONS Polymers as Surface-Based Tethers with Photolytic Triggers Enabling Laser-Induced Release/Desorption of Covalently Bound Molecules Jesus Ching,† Kamen I. Voivodov,†,‡ and T. William Hutchens*,† University of California, 110 FSTB, Davis, California 95616, and Molecular Analytical Systems, Inc., Suite 280, 2121 Sage Road, Houston, Texas 77056. Received April 15, 1996X
A novel design for suface-based macromolecular docking and release is presented together with a strategy to improve and extend biopolymer structure determination capabilities. Polymeric surfaces with arrays of tethers for covalent molecular attachment were designed with photolytic triggers to enable spatially defined, laser-induced uncoupling/desorption of the tethered molecules. Upon photolytic cleavage, a defined portion of the tether (“tail”) remains attached to the biomolecule as a probe. Chemically defined memory, determined by the number of reporter tails, reflects the biomolecule interaction with tether-probe devices encountered (i.e., footprint) on the probe surface. To demonstrate function, a surface of poly(4-vinylpyridine) was extended through the pyridinium nitrogens with spacer arms (-N-ethylsuccinamyl-) producing photolytic pyridinium nitrogen bonds. The photolabile tether was terminated with leaving groups (N-hydroxysulfosuccinimide) for covalent attachment of biopolymers. An 18-residue peptide (N terminus of human β-casein) was covalently docked to these tetherprobes, irradiated with coherent UV light, and released with two reporter tails of a mass predicted by tether formation at the two primary amine groups and subsequent photolytic cleavage at the intended site. This is the first demonstration of polymeric surface structure enabling the covalent docking and laser-induced uncoupling/desorption of intact macromolecules through the use of photolytic tethers. Surface-based tether-probe devices, operated by coherent light, should advance our ability to explore covalent modifications in biopolymer structure and alterations in conformation, generated either in advance of tethering or through chemical/enzymatic manipulations performed directly in situ.
A novel molecular design for surface-based macromolecular docking and release is presented together with a * Address correspondence to this author at 110 FSTB, University of California, Davis, CA 95616 [telephone (916) 7522662; facsimile (916) 752-5724; e-mail
[email protected]]. † University of California. ‡ Molecular Analytical Systems, Inc. X Abstract published in Advance ACS Abstracts, August 15, 1996.
S1043-1802(96)00042-0 CCC: $12.00
strategy to improve and extend biopolymer structure determination capabilities. The approach involves a new category of surface chemistries to covalently bind biopolymers that can later be uncoupled/desorbed simultaneously by pulsed laser irradiation. Pulsed laser irradiation has already been used to induce the desorption of structurally intact biopolymers from probes with chemically defined and molecular adsorption/desorption surface properties (1). The process © 1996 American Chemical Society
526 Bioconjugate Chem., Vol. 7, No. 5, 1996
Ching et al.
Figure 1. Modular design of biomolecular tether-probe devices. Each device is defined by three components attached to a specified surface architecture. A photolytic component (1) is placed at different relative positions (indicated by dotted arrows) within a variable length spacer arm (2) terminating with a bioconjugation component (3) for covalent attachment through specific sites on the target biomolecule surface. This device is designed so that laser irradiation induces a photolytic event that triggers uncoupling of the tethered biomolecule in a manner that leaves the dissociated biopolymer with one or more reporter tails. The detailed chemical structure of the device used to generate the data shown in Figure 2 is provided.
introduced in 1993, known as surface-enhanced laser desorption/ionization (SELDI) or affinity mass spectrometry (1, 2), was based on use of new probe material compositions and/or defined surface chemistries. In one category of process identified as surface-enhanced affinity capture (SEAC), chemically defined and biologically defined affinity capture devices are covalently bound directly to the modified probe surfaces. Only the captured target biomolecules, docked on the probe surface through noncovalent molecular recognition events, are selectively desorbed by laser irradiation. Inorganic and polymeric probe surfaces have been designed to present even covalently bound analyte biomolecules for laser-induced uncoupling/desorption (36). Our laboratory was able to accomplish, for the first time, the covalent attachment of biopolymers to a probe surface with tether chemistries enabling single pulses of laser irradiation to initiate subsequent uncoupling and desorption/ionization of the bound biomolecules (3, 4). This advance in laser-induced desorption capabilities revealed several new possibilities for the development of solid-phase bioconjugate chemistries for the investigation of biopolymer structure. One goal of this investigation was to determine whether the covalent binding of biomolecules to probe surfaces through more than one point of attachment could be achieved and still generate a homogeneous photolytic reaction product without loss of laser-induced uncoupling/desorption events that are, in effect, simultaneous. A further goal was to determine if polymers could be designed with the necessary photolabile biopolymer attachment and release characteristics. We have now explored a different approach to the design and synthesis of surfaces with the desired chemical reactivities and photolytic tethering capabilities. Polymeric surface structures with high-density arrays of chemically reactive groups for biopolymer attachment were required to have extremely efficient photolytic triggers to enable laser-induced release (uncoupling/ desorption) of the target biomolecules after covalent attachment. The covalent attachment/photolytic release process sought was conceived with the intent of deriving
polymer surface functionality not only to tether target biomolecules but also to probe biopolymer structure. We have now synthesized, within a single chemical construct, a combination tether-probe device and have tested its function both as a covalent photolabile tether and as a probe. A modular construction approach was adopted to optimize the contribution of three separate tether-probe device components designed to extend outward from a hydrophobic polymer surface (Figure 1). These components included a variable-length spacer arm, a photolytic component located at different positions within this spacer arm, and a zero-length bioconjugate component at the end of the spacer arm for covalent macromolecule attachment. The components within each device were intended to function collectively to accomplish several specific objectives. They must first act as a tether to covalently bind the target molecule at some defined distance from the probe surface. Second, after attachment, the photolytic component within the tether must remain an extremely efficient light-dependent triggering device for release of the tethered biomolecules. Finally, upon laser-induced uncoupling, a defined portion of the tether (“tail”) must be photochemically stable and remain covalently attached to the biomolecule as a probe. A simple procedure for the light-controlled release of large molecules from covalent anchor sites on a surface was dependent on our ability to demonstrate whether a single laser pulse could be used to induce simultaneously both uncoupling and desorption of covalently bound biomolecules (e.g., peptide) from a polymeric probe surface. More importantly, to implement the structural probe functions outlined, it was essential for us to determine whether simultaneous uncoupling/desorption could be accomplished in cases involving more than one covalent attachment site per macromolecule. To our knowledge, this has never been demonstrated. Several factors influenced our selection of a polymer as the foundation for the surface-based tether-probe device construction. Poly(vinylpyridine) is a nonconjugated hydrophobic polymer that has attracted attention
Communications
primarily as a charge-transfer catalyst with metal ions and as an electroactive polymer (7). For our purposes, however, the 4-vinylpyridine regioisomer (P4VP) presents sterically unhindered pyridyl nitrogen atoms in a ligand configuration which, upon aminoalkylation, should result in the introduction of a homogeneous class of photolytic bonds. Aminoalkylation also provides the chemical basis for extending outward with tether-probe construction. The photolytic site was incorporated at a defined position away from the biopolymer binding site by incorporation of a photoresistant spacer arm (-N-ethylsuccinamyl-). The spacer arm was terminated with a suitable leaving group (N-hydroxysulfosuccinimide) to achieve covalent attachment of the completed tether-probe device to the most reactive nucleophiles available at the target biopolymer surface. The model peptide chosen to evaluate tether-probe device functions, RETIESLSSSEESITEYK (R1-K18), was selected partly because it is a bioactive peptide containing multiple potential phosphorylation sites (consensus sequence) that define important transition metal ion binding properties; the precise structural determinants and chemical basis for specific metal ion binding are yet to be defined (8). Second, this peptide is produced from human β-casein by proteolytic cleavage at the C-terminal side of Lys (K18), a cleavage site common to all peptide fragments produced by tryptic digestion during peptide mapping experiments. Thus, R1-K18 represents other peptides resulting from trypsin cleavage in that it contains two likely covalent attachment sites, the R-amino group of the N terminus and the -amino group of the lysyl residue at the C terminus. The R1-K18 peptide was covalently bound to the tether-probe device illustrated in Figure 1. After the tether-probe device served its initial function as a tether, unbound R1-K18 peptide was removed and the second phase of the tether-probe function was initiated by laser irradiation. Laser-induced uncoupling and desorption of the covalently immobilized peptide were documented by time-of-flight mass analysis. These experiments revealed the release of a single macromolecular reaction product with an increase in mass equal to that predicted by a photolytic event occurring at the positively charged pyridinium nitrogen of each bound tether device (Figure 2, bottom). The mass increase observed for the bound peptide upon release (261 Da) was consistent with the formation of two reporter tails (127 Da each); one formed at each of the two available primary amino groups, as predicted. The tail mass of 127 Da was calculated without consideration of the probable addition of a hydrogen atom and/or proton. The precise mechanism of photo-induced bond cleavage at the pyridinium nitrogen is presently unknown. The 2-3-Da difference between the calculated and observed masses of the reporter tail may be due to a time delay or variation in the initial kinetic energy of the uncoupled peptide relative to that of the untethered internal mass standards. Regardless, the dissociated target macromolecule now has a chemically defined “memory” incorporated as a reflection of its interaction with the tether-probe devices on the biopolymer probe surface. No mass shift was evident for peptides not covalently bound to the tether-probe surface. Laser-induced desorption of the peptide presented on an unmodified polymer platform (i.e., not covalently bound) revealed a mass (2088.2 Da) equal to the mass calculated for this 18-residue peptide (Figure 2, top). The internal molecular mass standard was observed to have the correct molecular mass (2903.0 Da) and is shown as a reference point in both profiles.
Bioconjugate Chem., Vol. 7, No. 5, 1996 527
Figure 2. Laser-induced uncoupling, desorption, and mass analysis of the R1-K18 model peptide before (top) and after (bottom) interaction (i.e., covalent binding) and release from the biomolecular tether-probe device. Observed mass values for the protonated pseudomolecular ions [M + H]+ are shown. The mass observed for the R1-K18 model peptide after photolytic cleavage of the covalent tether(s) (2349 Da) is consistent with the generation of two reporter tails resulting from the formation of two tethers followed by photolytic cleavage of each at the site depicted in Figure 1. Peaks labeled [M + H]+ ) 2904 Da represent the internal mass calibration standard.
The efficiency and sensitivity of laser-induced desorption/ionization time-of-flight techniques (subfemtomole or attomole detection sensitivity) enable a much more detailed inspection of specific versus nonspecific proteinsurface interactions than was possible previously by any other method available. In the present case, it is clear that all traces of noncovalently bound biopolymer have been removed (see Figure 2). Although we have observed peptides uncoupled from this surface by a photolytic cleavage event involving the urea-type amide bond formed at the site of biomolecule docking, experimental factors associated with the formation of this so-called “0-tail” species are mostly associated with the use of other protocols for protein introduction to the surface. Biopolymer uncoupling/desorption data collected with other new types of tether-probe surface chemistries indicate the formation of tailed species and 0-tail species (5). Thus, surface-enhanced photolabile attachment and release (SEPAR) procedures may be modified or accomplished with surface chemistries constructed to allow macromolecular uncoupling/desorption with or without reporter tails. This capability should accommodate a variety of specific purposes and applications. The covalent docking of molecules, including biopolymers, in defined surface arrays facilitates remarkably the opportunity to address several broad categories of endeavor at the interface between chemistry and biology (1, 2, 9, 10). First, the ability to identify chemical and structural determinants of important molecular recogni-
528 Bioconjugate Chem., Vol. 7, No. 5, 1996
tion events is improved through covalent tethering approaches now typified by library screening (e.g., 8, 9). Our efforts to design photolabile bonds enabling simultaneously both covalent ligand uncoupling and desorption with laser irradiation, as an alternative approach to chemical-based cleavage and separation strategies, address objectives and anticipate needs such as those outlined recently by others developing support-bound combinatorial libraries (9, 10). A covalent docking approach also increases greatly the overall sensitivity and efficiency of both chemical and enzymatic digestion methods for the elucidation of biopolymer structure in situ (e.g., protein sequence and post-translational modifications). SEPAR probe functions to be derived from a surfacebased population of these devices may be anticipated in two ways. First, these functions display a collective ability to covalently anchor a target biopolymer through not just one, but several, attachment sites. This would enable enzymatic cleavage of the anchored biopolymer between attachment sites to facilitate subsequent structural evaluations, including internal sequence. Another application under evaluation is the potential of these and other similar tether-probe devices to be used as a tool to evaluate relative alterations in biopolymer size/shape or surface architecture. The light-dependent introduction of one or more well-defined chemical reporter groups (i.e., tails) to the target biomolecule resulting from its laserinduced departure from the surface may allow biopolymer conformations or “footprints” to be probed. Finally, the SEPAR process should, in principle, be applicable to the investigation of a wide variety of molecules, small and large, including not only proteins and other biopolymers such as oligonucleotides and polysaccharides but also synthetic polymers. ACKNOWLEDGMENT
We thank Hewlett-Packard for use of their laser desorption instrumentation. This work was supported by an NIH grant (R41GMS1658-01). Supporting Information Available: Experimental methods for the tether-probe device and spacer arm (1
Ching et al.
page). Ordering information is given on any current masthead page. LITERATURE CITED (1) Hutchens, T. W., and Yip, T. T. (1993) New Desorption Strategies for the Mass Spectrometric Analysis of Macromolecules. Rapid Commun. Mass Spectrom. 7, 576-580. (2) Hutchens, T. W., and Yip T. T. (1994) Affinity Mass Spectrometry: Probe Element Surfaces Enhanced for Affinity Capture (SEAC). Protein Sci. 3 (Suppl. 1), 81 (Abstract 170S). (3) Hutchens, T. W., and Yip, T. T. (1994) Method and Apparatus for Desorption and Ionization of Analytes. International Patent (PCT) Publication No. WO94/28418. (4) Hutchens, T. W., Ching, J., and Yip, T. T. (1995) Design of Mass Spectrometric Probe Surfaces for Photolabile Attachment and Release. Protein Sci. 4 (Suppl. 2), 99 (Abstract 204S) . (5) Ching, J., Voivodov, K. I., and Hutchens, T. W. (1996) Surface Chemistries Enabling Photo-Induced Uncoupling/ Desorption of Covalently Tethered Biomolecules. J. Org. Chem. 61, 3582-3583. (6) Voivodov, K. I., Ching, J., and Hutchens, T. W. (1996) Surface Arrays of Energy Absorbing Polymers Enabling Covalent Attachment of Biomolecules for Subsequent LaserInduced Uncoupling/Desorption. Tetrahedron Lett. 37, 56695672. (7) Greg, B. A., and Heller, A. (1991) Redox Polymer Films Containing Enzymes. 1. A Redox-Conducting Epoxy Cement: Synthesis, Characterization, and Electrocatalytic Oxidation of Hydroquinone. J. Phys. Chem. 95, 5970-5975. (8) Hutchens, T. W., and Yip, T. T. (1992) Model protein surface domains for the investigation of metal ion-dependent macromolecular interactions and metal ion transfer. Methods (San Diego) 4, 79-96. (9) Egner, B. J., Langley, J. G., and Bradley, M. (1995) Solid Phase Chemistry: Direct Monitoring by Matrix-Assisted Laser Desorption/Ionization Time of Flight Mass Spectrometry. A Tool for Combinatorial Chemistry. J. Org. Chem. 60, 2652-2653. (10) Brummel, C. L., Lee, I. N. W., Zhou, Y., Benkovic, S. J., and Winograd, N. (1994) A Mass Spectrometric Solution to the Address Problem of Combinatorial Libraries. Science 264, 399-401.
BC9600426
