Rapid peptide fragmentation without electrons, collisions, infrared radiation, or native chromophores

Geoffrey K. Yeh 0 1 2 Qingyu Sun 0 1 2 Claudia Meneses 0 1 2 Ryan R. Julian 0 1 2 0 Address reprint requests to Dr. R. R. Juliian, Department of Chemistry, University of California , Riverside, Riverside, CA 92508, USA 1 Published online November 5, 2008 Received August 26, 2008 Revised October 28, 2008 Accepted October 28, 2008 2 Department of Chemistry, University of California at Riverside , Riverside, California, USA Ultraviolet photodissociation of peptides followed by mass analysis has several desirable advantages relative to other methods, yet it has not found widespread use due to several limitations. One shortcoming is the inefficiency with which peptides absorb in the ultraviolet. This issue has a simple solution and can be circumvented by the attachment of noncovalent adducts that contain appropriate chromophores. Subsequent photoactivation of the chromophore leads to vibrational excitation of the complex and eventually to fragmentation of the peptide. Herein, the energetics that control the efficiency of this process are examined as a function of the characteristics of both the peptide and the noncovalently attached chromophore. Fragmentation efficiency decreases with increasing peptide size and is also constrained by the binding energy of the noncovalent adduct. The optimum chromophore should have excellent absorption at the excitation wavelength and a low luminescence quantum yield. It is demonstrated that a naphthyl based 18-crown-6 adduct is ideally suited for attaching to a variety peptides and fragmenting them following absorption of 266 nm light. Potential applications and limitations of this methodology are discussed. (J Am Soc Mass Spectrom 2009, 20, 385-393) 2009 Published by Elsevier Inc. on behalf of American Society for Mass Spectrometry - trast, ultraviolet photodissociation (UVPD) occurs on a much shorter time scale and can yield fragment ions characteristic of CID or IRMPD. However, the major limitation of UVPD is that the precursor ion must contain a suitable chromophore. For peptides, UVPD requires either short wavelength lasers (less than 200 nm) or peptides rich in tyrosine, tryptophan, and phenylalanine that can absorb at longer wavelengths. Therefore, UVPD is either limited to a small subset of all peptides or constrained by undesirable technical challenges which are encountered when working in the vacuum ultraviolet. Thus, widespread implementation of UVPD for peptide fragmentation has been to date somewhat limited. However, Brodbelt and coworkers have recently described a method to circumvent the requirement for aromatic residues or short wavelength lasers [12]. In this work, a noncovalently attached chromophore was used to absorb energy from a photon and transfer it to a peptide, causing the peptide to fragment. 18-crown-6 ether (18C6), which preferentially associates with lysine residues via the formation of three hydrogen bonds, was used for the noncovalent attachment [1315]. When combined with a dipyrrolylquinoxaline chromophore, the crown/peptide interaction was found to accommodate vibrational energy transfer from the chromophore to the peptide. In this manner, energy can be transferred from the laser to the chromophore, and then from the chromophore to the peptide (following internal converTmethodologies available for interrogating pephere are a variety of mainstream fragmentation tides, including collision-induced dissociation (CID) [1], infrared multiphoton photodissociation (IRMPD) [2], electron capture dissociation (ECD) [3], and electron-transfer dissociation (ETD) [4]. In addition, there are other methods that can be used, such as surface induced dissociation (SID) [5], black-body infrared radiative dissociation (BIRD) [6], along with various photodissociation experiments employing high-energy lasers [711]. Though none of these techniques is perfect, each has characteristic strengths and weaknesses. For example CID is easily employed in the source region of any instrument with atmospheric pressure ionization. CID is also easily accommodated elsewhere in an instrument as long as high vacuum is not required. In contrast, IRMPD is well suited for collisionless environments. In this regard, CID and IRMPD provide complementary capabilities. Furthermore, since both methods fragment ions by the stepwise addition of small amounts of energy, the results obtained are frequently quite similar. However, CID and IRMPD are both limited by the fact that energy deposition occurs over a relatively long time scale, typically milliseconds to seconds. In consion into vibrational energy). Following excitation of the peptide, the crown adduct departs and the peptide fragments to yield products similar to those observed by CID. Herein we report the discovery of a naphthyl based 18C6 reagent (NC), which behaves similarly. Excitation of peptide/NC complexes results in efficient energytransfer to the peptide followed by fragmentation of the peptide. Importantly for NC adducts, a single laser pulse is found to be sufficient to induce fragmentation, meaning that CID like spectra can be obtained in a collisionless environment on a microsecond time scale. Further experiments were conducted to explore the issues that influence the degree of peptide fragmentation efficiency. The following factors were found to be important: the nature of the chromophore, the number of degrees of freedom in the peptide (or peptide size), and the energy gap between the barriers for noncovalent disassembly and covalent bond dissociation. Ideally, the chromophore should exhibit strong absorption and internal conversion. Fragmentation decreases as peptide size increases and as the amount of energy required for dissociating the peptide begins to significantly exceed the noncovalent binding energy. Nevertheless, under appropriate conditions peptides comprising a wide range of sizes and displaying substantial sequence diversity can be fragmented. This includes peptides that do not contain lysine and peptides exceeding 2000 Da. Furthermore, it is demonstrated that a single PD spectrum frequently contains fragments from multiple charge states, meaning that high information content is revealed in each spectrum. Materials and Methods KP13K was synthesized by Genscript (Piscataway, NJ). Melittin, kassinin, Lys [0]bradykinin, bradykinin, dynorphin- , GSK- inhibitor, myosin kinase inhibitor, and mouse ANP [1-11] were purchased from American Peptide Company (Sunnyvale, CA). Polyalanine peptides were synthesized in-house using standard Fmoc coupling methodology. Water was purified in-house to a minimum resistance of 18 M . HPLC grade acetonitrile (ACN), ACS grade anhydrous tetrahydrofuran (THF), and ACS grade triethylamine (TEA) were purchased from Sigma Aldrich (St. Louis, MO). NC and dansyl-crown (DC) were prepared by dissolving 2-hydroxymethyl-18-crown-6 (Sigma-Aldrich) in anhydrous THF containing a catalytic amount of triethylamine. 2-Napthoyl chloride or dansyl chloride (...truncated)