Building a research toolkit for protein tyrosine phosphatases

Protein tyrosine phosphatases (PTPs) play critical roles in cell signaling processes. The aberrant activities of PTPs are highly associated with many diseases such as type-1 diabetes, rheumatoid arthritis, and cancer. Despite their importance, the biological roles and regulation of individual PTPs are poorly understood due to the shortage of sufficient detection methods. To facilitate PTP research, two approaches were used to develop selective, peptide-based PTP chemical probes. The first approach was a rational design method, in which we developed a SHP2- selective fluorescent substrate based on the physiological PTP substrate TIE2 with a high micromolar affinity (KM = 0.7 ± 0.2 mM) towards SHP2. We further optimized the probe based on substrate sequence information available in the literature, which resulted in a 7- fold increase in reactivity. The second approach involved the use of a combinatorial substrate library, in which we designed, built, and screened a diverse library of peptides using an inverse alanine screening approach and obtained substrate positional preference profiles for 11 PTPs at 8 positions surrounding the phosphotyrosine. Based on the preference profile of LMWPTP and PTPμ, we developed two fluorescent probes and achieved up to a 6-fold increase on reactivity compared to a negative control substrate. The preference profile can be used for future probe development and optimization for the 11 PTPs. In addition to fluorescent substrates, we also developed two activity-based probes using a similar approach by incorporating a covalent inhibitor, 4-(azidomethyl) iv phenylvinyl sulfonate, at the phosphotyrosine position of the peptide sequence. Both probes exhibited time- and concentration-dependent inhibitory activity towards PTPN22 and SHP2. The EDNE probe has an IC50 of 130 ± 6 μM towards SHP2 after a one-hour incubation, while the LDLL probe has an IC50 of 80 ± 10 μM under the same condition. The EDNE probe exhibits up to 32-fold selectivity over PTP1B, PTP-PEST, TCPTP, and YopH. Furthermore, the EDNE probe shows selectivity towards SHP2 over other cysteine-based enzymes in a SHP2-expressing lysate. Both the substrates and inhibitors we developed add to the toolkit of chemical probes used to help better understand the regulation and physiological roles of PTPs.

BUILDING A RESEARCH TOOLKIT FOR PROTEIN TYROSINE PHOSPHATASES by Shuangyu Ma A dissertation submitted to the faculty of The University of Utah in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Medicinal Chemistry The University of Utah May 2019 Copyright © Shuangyu Ma 2019 All Rights Reserved The University of Utah Graduate School STATEMENT OF DISSERTATION APPROVAL The dissertation of Shuangyu Ma has been approved by the following supervisory committee members: Amy Marie Barrios , Chair 2/22/2019 Date Approved Eric W. Schmidt , Member 2/22/2019 Date Approved Darrell R. Davis , Member 2/22/2019 Date Approved Kuberan Balagurunathan , Member 2/25/2019 Date Approved Michael S. Kay , Member 2/22/2019 Date Approved and by the Department/College/School of Darrell R. Davis , Chair/Dean of Medicinal Chemistry and by David B. Kieda, Dean of The Graduate School. ABSTRACT Protein tyrosine phosphatases (PTPs) play critical roles in cell signaling processes. The aberrant activities of PTPs are highly associated with many diseases such as type-1 diabetes, rheumatoid arthritis, and cancer. Despite their importance, the biological roles and regulation of individual PTPs are poorly understood due to the shortage of sufficient detection methods. To facilitate PTP research, two approaches were used to develop selective, peptide-based PTP chemical probes. The first approach was a rational design method, in which we developed a SHP2selective fluorescent substrate based on the physiological PTP substrate TIE2 with a high micromolar affinity (KM = 0.7 ± 0.2 mM) towards SHP2. We further optimized the probe based on substrate sequence information available in the literature, which resulted in a 7fold increase in reactivity. The second approach involved the use of a combinatorial substrate library, in which we designed, built, and screened a diverse library of peptides using an inverse alanine screening approach and obtained substrate positional preference profiles for 11 PTPs at 8 positions surrounding the phosphotyrosine. Based on the preference profile of LMWPTP and PTPµ, we developed two fluorescent probes and achieved up to a 6-fold increase on reactivity compared to a negative control substrate. The preference profile can be used for future probe development and optimization for the 11 PTPs. In addition to fluorescent substrates, we also developed two activity-based probes using a similar approach by incorporating a covalent inhibitor, 4-(azidomethyl) phenylvinyl sulfonate, at the phosphotyrosine position of the peptide sequence. Both probes exhibited time- and concentration-dependent inhibitory activity towards PTPN22 and SHP2. The EDNE probe has an IC50 of 130 ± 6 µM towards SHP2 after a one-hour incubation, while the LDLL probe has an IC50 of 80 ± 10 µM under the same condition. The EDNE probe exhibits up to 32-fold selectivity over PTP1B, PTP-PEST, TCPTP, and YopH. Furthermore, the EDNE probe shows selectivity towards SHP2 over other cysteine-based enzymes in a SHP2-expressing lysate. Both the substrates and inhibitors we developed add to the toolkit of chemical probes used to help better understand the regulation and physiological roles of PTPs. iv TABLE OF CONTENTS ABSTRACT....................................................................................................................... iii LIST OF TABLES ............................................................................................................ vii ACKNOWLEDGEMENTS ............................................................................................. viii Chapters 1. INTRODUCTION .......................................................................................................... 1 1.1 Protein tyrosine phosphorylation and the protein tyrosine phosphatase superfamily .......................................................................................................... 1 1.2 The catalytic mechanism of PTPs ........................................................................ 3 1.3 PTPs as intriguing therapeutic targets.................................................................. 3 1.4 Redox regulation of PTPs .................................................................................... 5 1.5 Targeting PTPs remains challenging ................................................................... 6 1.6 Reference ............................................................................................................. 7 2. RATIONAL DESIGN OF SHP2-SPECIFIC FLUOROGENIC SUBSTRATES AS MECHANISM-BASED PROBES ............................................................................. 14 2.1 Introduction ........................................................................................................ 14 2.2 Results and discussion ....................................................................................... 16 2.3 Conclusion ......................................................................................................... 18 2.4 Experimental section .......................................................................................... 19 2.5 References .......................................................................................................... 25 3. SYNTHETIC PEPTIDE LIBRARY SCREENING FOR PTPS .................................. 34 3.1 Introduction ........................................................................................................ 34 3.2 Results and discussion ....................................................................................... 36 3.3 Conclusion ......................................................................................................... 41 3.4 Experiments and methods .................................................................................. 42 3.5 Reference ........................................................................................................... 45 4. ACTIVITY-BASED PEPTIDE PROBES FOR TARGETING SPECIFIC PTPS ....... 61 4.1 Introduction ........................................................................................................ 61 4.2 Results and discussion ....................................................................................... 62 4.3 Conclusion ......................................................................................................... 67 4.4 Experimental section .......................................................................................... 68 4.5 Reference ........................................................................................................... 74 5. DUAL COLORIMETRIC AND FLUOROGENIC PROBES FOR VISUALIZING TYROSINE PHOSPHATASE ACTIVITY................................................................ 86 5.1 Notes and references .......................................................................................... 89 vi LIST OF TABLES Tables 2.1 Three pCAP peptides with different sequences and reactivities towards TCPTP ...... 28 3.1 A summary of the internalization efficiencies and toxicities of the CPP library........ 47 3.2 Library screen data V0, color coded ............................................................................ 52 3.3 Library screen data normalized to local maximum, color coded. ............................... 55 3.4 A proposed pool of lead sequences ............................................................................. 58 3.5 The sequence of PTPµ probe and LMWPTP probe ................................................... 59 3.6 The final yields of the purified CPP library.. .............................................................. 60 4.1 The selectivity of EDNE probe and LDLL probe.. ..................................................... 78 4.2 Summary of the SHP2 trypsin digestion/probe conjugate mapping experiment result............................................................................................................................ 85 5.1 Advantages and limitations of selected PTP substrates.. ............................................ 88 5.2 Kinetic constants for PTP assays using pRes as the substrate.. .................................. 88 5.3 Kinetic constants for PTP assays using F2pRes as the substrate.. .............................. 89 ACKNOWLEDGEMENTS First of all, I would like to thank my Ph.D. advisor Dr. Amy M. Barrios. Over the past five years, she has always been a great mentor not only for giving me excellent scientific training, but also for supporting me in developing and working towards longterm career goals. With her support, I learned many technical skills, and had various opportunities to become better at critical thinking, gain the confidence to face and solve challenges, and learn how to collaborate with people and lead projects. Secondly, I would like to thank all my colleagues from the Barrios lab, especially Brandon McCullough and Dr. Suvendu Biswas. Many of the experimental plans and analyses were inspired from thorough discussions with them. I would also like to thank Dr. Andrew C. B. Cato and Dr. George Ainooson for being the most considerate collaborators during my visiting research trip in Germany. In addition, I would like to thank my parents. They have supported me to follow my passion and pursue a graduate degree abroad. I would also like to thank my friends in Salt Lake City for keeping me accompanied when I was not in the lab. Last but not the least, I would like to thank the College of Pharmacy, the National Science Foundation, and the Skaggs Graduate Research Fellowship program. Without their support, none of these research projects would have been possible. CHAPTER 1 INTRODUCTION 1.1 Protein tyrosine phosphorylation and the protein tyrosine phosphatase superfamily Protein tyrosine phosphorylation is one of the most important post-translational modifications in the cell. It regulates a broad range of fundamental cellular functions, including cell growth, differentiation, communication, migration, immune response, etc.1 Among the 30% of all cellular proteins being phosphorylated,2 less than 1% of the phosphorylated amino acid residues are phosphotyrosines.3 Despite the low abundance of tyrosine phosphorylation, this post-translational modification often plays central roles in regulating signaling pathways,2 such as growth factor receptor signaling and integrin signaling, as well as other cellular processes such as neural transmission.1 The phosphorylation and dephosphorylation of key signaling proteins are tightly regulated by two families of enzymes: the protein tyrosine kinases (PTKs) which phosphorylate the tyrosine residues of the substrate protein in an ATP-dependent manner, and the protein tyrosine phosphatases (PTPs) which catalyze the reverse process. The PTKs have been extensively studied since 1979.1 Because of their positive regulatory role in many signaling pathways, PTKs are regarded as popular therapeutic targets. Many PTK inhibitors have been developed into drugs, such as sunitinib for renal carcinoma treatment, or lapatinib for breast cancer therapy.1 2 PTPs, however, are much less understood compared to the PTKs. There was significant debate about whether the removal of the phosphate group was catalyzed by an enzyme until the first PTP was discovered by Tonks et al in 1988,4 almost ten years after the first discovery of PTKs.5 Since then, it has been conclusively demonstrated that the phosphate removal process is actively controlled by PTPs.6 The PTP family, much like the PTK family, is well regulated in response to various cellular signals, such as growth factors and cytokines. Tonks et al also identified the signature catalytic motif of PTPs, HC(X)5R, which is conserved across the entire superfamily. There are 107 PTP domains encoded in the human genome that have this signature motif, indicating that the PTP and PTK enzyme families are similar in size (107 PTPs as compared to 90 PTKs in the human genome).1 The 107 members in the PTP superfamily are further categorized into 4 subfamilies according to their structural and functional differences: i) the classical pTyrspecific PTPs, ii) the dual specificity phosphatases, iii) Cdc25 phosphatases, and iv) low molecular weight PTP (LMWPTP) (Figure 1.1).7 The classical pTyr-specific PTPs dephosphorylate only pTyr, while the dual specificity phosphatases can hydrolyze pTyr as well as pSer/pThr. Cdc25 phosphatases also exhibit dual specificity towards the phosphorylated substrates, but they are more distantly related to other members of the PTP superfamily at both primary and tertiary structural levels. Except for the LMWPTP subfamily, PTPs usually have different compositions of regulatory sub-domains, but they all have at least one PTP domain which catalyzes the tyrosine dephosphorylation reaction. The LMWPTP subfamily consists of two isoforms of one member, both of which only have the PTP domain.8 Some PTPs, such as PTPδ and CD45, have an additional PTP domain which is inactive and often provides a negative regulatory 3 function.9 1.2 The catalytic mechanism of PTPs Although PTPs have a diverse range of substrates and various regulatory roles in signaling events, the same catalytic mechanism is shared across the entire family. This is largely due to a high sequence conservation at the PTP catalytic site (Figure 1.2).7 They all contain a P-loop, which is responsible for the phosphate recognition and binding. The catalytic cysteine residue, which is completely invariant, is located in the P-loop and performs the nucleophilic attack on the phosphorus. An aspartate residue, which is also invariant, is located in the flexible WPD-loop to assist in substrate binding, acting as a general acid (and general base in the later stage) to facilitate the phosphate hydrolysis.10 In addition, a conserved glutamine residue in the Q-loop places a nucleophilic water to help hydrolyze the enzyme-phosphate intermediate and restore the catalytic cysteine in as the final step. 1.3 PTPs as intriguing therapeutic targets Due to its crucial physiological role, aberrant PTP activity is involved in many human diseases, including cancer,1 systemic lupus erythematosus,11 obesity, and diabetes.12 For example, hematopoietic tyrosine phosphatase (HePTP) regulates the activity and translocation of the MAP kinases Erk and p38; it is highly expressed in acute myeloid leukemia and the expression in fibroblasts resulted in transformation.13 Another example is PTP1B, which negatively regulates insulin signaling as well as hypothalamic leptin signaling.14,15 As a result, mice lacking PTP1B are resistant to high-fat diet-induced obesity.16 Lymphoid-specific tyrosine phosphatase (PTPN22) is a negative regulator of 4 T-cell receptor signaling pathway. A single nucleotide polymorphism has been correlated with a significant increase in the risk of autoimmune diseases, including type-1 diabetes, juvenile rheumatoid arthritis, and systemic lupus erythematosus.17,18 Since many activated PTKs are identified as oncoproteins, PTPs are usually regarded as tumor suppressors.19 For example, PTEN suppresses tumor growth by downregulating the PI3K-dependent signaling pathways that are associated with cell survival.20 Deletions and mutations of DEP1, which dephosphorylates and inhibits the MAPK pathway, were found in multiple cancers including colon, lung, and breast cancer.21,22 However, a prominent exception to this norm is SHP2, which was identified as the first oncoprotein in this superfamily. SHP2 has two auto-inhibitory SH2 domains, a PTP domain, and a carboxyl-terminal tail. In resting cells, SHP2 is auto-inhibited through intramolecular interactions between its N-terminal SH2 domain and the PTP active site.23 Upon stimulation by growth factors or cytokines, an upstream signaling protein such as Gab1/Gab2 activates SHP2 by binding with SH2 domain, which removes the autoinhibition and restores the activity of the PTP domain.24 Recent studies have shown SHP2 to be a key positive regulator of the Ras-Raf-ERK and PI3K-AKT pathways, both of which are critical oncogenic pathways supporting cancer cell survival and proliferation.24,25 Mutations in SHP2 are believed to cause Noonan syndrome (activating mutations) and LEOPARD syndrome (inactivating mutations).26,27 Elevated SHP2 activities are found in many different types of malignancies.28 In addition, SHP2 is a key mediator in the T-cell receptor signaling pathway and is regarded as a potential target for tumor immunotherapy.29 Based on these findings, SHP2 has become a popular target in recent years for new anti-cancer therapy development. 5 1.4 Redox regulation of PTPs Inhibition of PTP activity through reversible oxidation of the catalytic cysteine residue has attracted a lot of interest as a biological regulatory mechanism of these enzymes in recent years.30 Due to the microenvironment in the active site, the catalytic cysteine residue of the PTPs has a perturbed pKa value (~4.5) which makes it prone to oxidation.30 Researchers have demonstrated that the inhibitory oxidation of PTPs can be induced by reactive oxygen species (ROS), which are produced in response to a variety of stimuli, including growth factors, cytokine, and UV light.31 The initially formed sulfenic acid product can form secondary reaction products rapidly by reacting with a neighboring peptide backbone to yield a sulphenylamide, or with another cysteine in close proximity to yield a disulfide.32 These secondary oxidation products are reversible in the presence of reductants, thus protecting the catalytic cysteine from irreversible oxidation (Figure 1.3). Increasing evidence correlates PTP oxidation with alterations of PTP proteinprotein interactions as well as misregulation in cell signaling.30 For example, oxidization of PTP1B can enhance its binding with calpain and result in facilitated proteolytic degradation.33 In another example, after the deletion of p66Shc, a protein that produces mitochondrial ROS, the decreased PTP oxidation was observed which is then related to the attenuation of the biochemical and cellular responses to PDGF stimulation.30 PTP oxidation is also correlated with multiple pathologies such as cancer,34 insulin resistance,35 and hypoxia-reoxygenation.36 However, direct causal relationships between these pathologies and PTP oxidation have not been established. 6 1.5 Targeting PTPs remains challenging Despite the large number of PTPs in the human genome and their underlying physiological importance, PTPs have not been successfully targeted therapeutically, and their regulatory mechanisms are poorly understood.37 For decades, the research on PTPs has lagged behind that of the PTKs.6 A major challenge that impedes the development of PTP-targeted chemical probes is the structural similarities of the PTPs, especially in the catalytic domain.38 Due to the high sequence conservation, existing small molecule probes are often pan-specific to multiple PTPs.7 Many commercially available antibodies also failed to achieve good specificity towards specific PTPs. To further complicate the problem, the catalytic pocket in the PTP domain is often positively charged, which causes it to prefer acidic ligands.39 As a result, library screening aimed at identifying PTP probes or inhibitors often yields highly acidic leads with poor cell permeability and low bioavailability.38 Therefore, the discovery of new research tools with good specificity and bioavailability is very much needed to help facilitate our understanding of PTPs, and to eventually target PTPs for therapeutic benefit. To develop useful tools for PTP biological research, we designed a series of chemical probes through a combination of rational design and library screenings. In Chapter 2, we synthesized a list of SHP2-selective fluorescent probes based on natural substrates from the literature, and validated the principle of using peptide-based substrates to achieve selectivity on certain PTPs over the others. To improve the cell permeability of the peptide-based probes and guide probe design for PTPs with limited knowledge of their natural substrate, we took a library screening approach as described in Chapter 3 to obtain positional preference profiles of 11 PTPs, which can be used to guide the design of selective PTP sequences that tolerate a 7 highly basic, polyarginine cell-penetrating tag. In addition to fluorescent substrates, we designed a selective covalent inhibitor for SHP2 (Chapter 4), which can also serve as an affinity handle to pull down SHP2 from complex biological samples in the presence of other cysteine containing proteins and enzymes. Finally, in Chapter 5, the development of a red-shifted fluorescent PTP probe that can be used in live cell imaging and potentially be developed into a next-generation warhead molecule for future peptide-based probe design is described. With all these tools we developed, our final goal is to pave the way for PTP biological studies and eventually help complete our understanding of PTPs in multiple signaling pathways. 1.6 Reference (1) Hunter, T. Curr. Opin. Cell Biol. 2009, 21, 140. (2) Ubersax, J. A.; Ferrell, J. E. Nat. Rev. Mol. Cell Biol. 2007, 8, 530. (3) Hunter, T. Cold Spring Harb. Perspect. Biol. 2014, 6, 1. (4) Tonks, N. K.; Diltz, C. D.; Fischer, E. H. J. Biol. Chem. 1988, 263, 6731. (5) Hunter, T. Proc. Natl. Acad. Sci. 2015, 112, 7877. (6) Tonks, N. K. FEBS J. 2013, 280, 346. (7) Zhang, Z.-Y. Acc. Chem. Res. 2003, 36, 385. (8) Caselli, A.; Paoli, P.; Santi, A.; Mugnaioni, C.; Toti, A.; Camici, G.; Cirri, P. Biochim. Biophys. Acta - Proteins Proteomics 2016, 1864, 1339. (9) Wallace, M. J.; Fladd, C.; Batt, J.; Rotin, D. Mol. Cell. Biol. 1998, 18, 2608. (10) Zhao, S.; Sedwick, D.; Wang, Z. Genetic alterations of protein tyrosine phosphatases in human cancers. Oncogene 2015, 34, 3885–3894. (11) Fousteri, G.; Liossis, S.-N. C.; Battaglia, M. Clin. Immunol. 2013, 149, 556. (12) Knobler, H.; Elson, A. J. Biomed. Res. 2014, 28, 157. (13) Pettiford, S. M.; Herbst, R. Oncogene 2000, 19, 858. 8 (14) Zabolotny, J. M.; Bence-Hanulec, K. K.; Stricker-Krongrad, A.; Haj, F.; Wang, Y.; Minokoshi, Y.; Kim, Y. B.; Elmquist, J. K.; Tartaglia, L. A.; Kahn, B. B.; Neel, B. G. Dev. Cell 2002, 2, 489. (15) Shimizu, S.; Kolls, J. K.; Bryer-Ash, M.; Nishio, Y.; Cheung, A. T.; Kashiwagi, A.; Kikkawa, R.; Egawa, K.; Maegawa, H.; Morino, K. J. Biol. Chem. 2002, 276, 10207. (16) Bence, K. K.; Delibegovic, M.; Xue, B.; Gorgun, C. Z.; Hotamisligil, G. S.; Neel, B. G.; Kahn, B. B. Nat. Med. 2006, 12, 917. (17) Bottini, N.; Vang, T.; Cucca, F.; Mustelin, T. Semin. Immunol. 2006, 18, 207. (18) Gregersen, P. K. Nat Genet. 2005, 37, 1300. (19) Scott, A.; Wang, Z. Biosci. Rep. 2011, 31, 303. (20) Di Cristofano, A.; Pesce, B.; Cordon-Cardo, C.; Pandolfi, P. P. Nat. Genet. 1998, 19, 348. (21) Tonks, N. K. Nat. Rev. Mol. Cell Biol. 2006, 7, 833. (22) Du, Y.; Grandis, J. R. Receptor-type protein tyrosine phosphatases in cancer. Chinese Journal of Cancer 2014, 34, 61–69. (23) Hof, P.; Pluskey, S.; Dhe-Paganon, S.; Eck, M. J.; Shoelson, S. E.; Allard, J.; Chang, H.; Herbst, R.; McNeill, H.; Simon, M.; Arrandale, J.; Gore-Willse, A.; Rocks, S.; Ren, J.; Zhu, J.; Davis, A.; Livingston, J.; Rabin, D.; Backer, J.; Myers, M.; Shoelson, S.; Chin, D.; Sun, X.; Miralpeix, M.; Hu, P.; Margolis, B.; Skolnik, E.; Schlessinger, J.; White, M.; Baldwin, E.; Xu, J.; Hajiseyedjavadi, O.; Baase, W.; Matthews, B.; Barford, D.; Flint, A.; Tonks, N.; Bennett, A.; Hausdorff, S.; O’Reilly, A.; Freeman, R.; Neel, B.; Bilwes, A.; Hertog, J. den; Hunter, T.; Noel, J.; Brunger, A.; Carpenter, C.; Auger, K.; Chaudhuri, M.; Yoakim, M.; Schaffhausen, B.; Shoelson, S.; Cantley, L.; Carson, M.; Cohen, G.; Ren, R.; Baltimore, D.; 4, C. C. P.; Dechert, U.; Adam, M.; Harder, K.; Clark-Lewis, I.; Jirik, F.; Eck, M.; Pluskey, S.; Trub, T.; Harrison, S.; Shoelson, S.; Eriksson, A.; Baase, W.; Zhang, X.; Heinz, D.; Blaber, M.; Baldwin, E.; Matthews, B.; Feng, G.; Hui, C.-C.; Pawson, T.; Fujioka, Y.; Matozaki, T.; Noguchi, T.; Iwamatsu, A.; Yamao, T.; Takahashi, N.; Tsuda, M.; Takada, T.; Kasuga, M.; Hoffmann, K.; Tonks, N.; Barford, D.; Jackson, S.; elMasry, N.; Fersht, A.; Jackson, S.; Moracci, M.; elMasry, N.; Johnson, C.; Fersht, A.; Jackson, D.; Kupcho, K.; Newman, P.; Jia, Z.; Barford, D.; Flint, A.; Tonks, N.; Kabsch, W.; Kharitonenkov, A.; Chen, Z.; Sures, I.; Wang, H.; Schilling, J.; Ullrich, A.; Kraulis, P.; Kuhne, M.; Pawson, T.; Lienhard, G.; Feng, G.; Kuriyan, J.; Cowburn, D.; Lamzin, V.; Wilson, K.; Lechleider, R.; Sugimoto, S.; Bennett, A.; Kashishian, A.; Cooper, J.; Shoelson, S.; Walsh, C.; Neel, B.; Lee, C.-H.; Kominos, D.; Jaques, S.; Margolis, B.; Schlessinger, J.; Shoelson, S.; Kuriyan, J.; Neel, B.; Tonks, N.; Nicholls, A.; Sharp, K.; Honig, B.; O’Reilly, A.; Neel, B.; Ottinger, E.; Botfield, M.; Shoelson, S.; Pawson, T.; Pei, D.; Lorenz, U.; Klingmuller, U.; Neel, B.; Walsh, C.; Pei, D.; 9 Wang, J.; Walsh, C.; Perkins, L.; Larsen, I.; Perrimon, N.; Pluskey, S.; Wandless, T.; Walsh, C.; Shoelson, S.; Rhee, S.; Bae, Y.; Richards, F.; Kundrot, C.; Rowley, R.; Burkhardt, A.; Chao, H.; Matsueda, G.; Bolen, J.; Saxton, T.; Henkemeyer, M.; Gasca, S.; Shen, R.; Rossi, D.; Shalaby, F.; Feng, G.; Pawson, T.; Shiue, L.; Zoller, M.; Brugge, J.; Sicheri, F.; Moarefi, I.; Kuriyan, J.; Stuckey, J.; Schubert, H.; Fauman, E.; Zhang, Z.-Y.; Dixon, J.; Saper, M.; Sugimoto, S.; Lechleider, R.; Shoelson, S.; Neel, B.; Walsh, C.; Sugimoto, S.; Wandless, T.; Shoelson, S.; Neel, B.; Walsh, C.; Tang, T.; Freeman, R.; O’Reilly, A.; Neel, B.; Sokol, S.; Tonks, N.; Townley, R.; Shen, S.-H.; Banville, D.; Ramachandran, C.; Vogel, W.; Lammers, R.; Huang, J.; Ullrich, A.; Welham, M.; Dechert, U.; Leslie, K.; Jirik, F.; Schrader, J.; Williams, J.; Weijland, A.; Gonfloni, S.; Thompson, A.; Courtneidge, S.; Superti-Furga, G.; Wierenga, R.; Xu, W.; Harrison, S.; Eck, M. Cell 1998, 92, 441. (24) Zhang, J.; Zhang, F.; Niu, R. J. Cell. Mol. Med. 2015, 19, 2075. (25) Kano, Y.; Cook, J. D.; Lee, J. E.; Ohh, M. Semin. Cell Dev. Biol. 2016, 58, 70. (26) Noda, S.; Takahashi, A.; Hayashi, T.; Tanuma, S. I.; Hatakeyama, M. Biochem. Biophys. Res. Commun. 2016, 469, 1133. (27) Tartaglia, M.; Mehler, E. L.; Goldberg, R.; Zampino, G.; Brunner, H. G.; Kremer, H.; van der Burgt, I.; Crosby, a H.; Ion, A.; Jeffery, S.; Kalidas, K.; Patton, M. a; Kucherlapati, R. S.; Gelb, B. D. Nat. Genet. 2001, 29, 465. (28) Ran, H.; Tsutsumi, R.; Araki, T.; Neel, B. G. Cancer Cell 2016, 30, 194. (29) Lorenz, U. Immunol Rev. 2009, 228, 342. (30) Östman, A.; Frijhoff, J.; Sandin, A.; Böhmer, F. D. J. Biochem. 2011, 150, 345. (31) Chen, C. Y.; Willard, D.; Rudolph, J. Biochemistry 2009, 48, 1399. (32) Den Hertog, J.; Groen, A.; Van Der Wijk, T. Arch. Biochem. Biophys. 2005, 434, 11. (33) Gulati, P.; Markova, B.; Göttlicher, M.; Böhmer, F. D.; Herrlich, P. A. EMBO Rep. 2004, 5, 812. (34) Leonard, S. E.; Garcia, F. J.; Goodsell, D. S.; Carroll, K. S. Angew. Chemie Int. Ed. 2011, 50, 4423. (35) Lou, Y. W.; Chen, Y. Y.; Hsu, S. F.; Chen, R. K.; Lee, C. L.; Khoo, K. H.; Tonks, N. K.; Meng, T. C. FEBS J. 2008, 275, 69. (36) Sandin, Å.; Dagnell, M.; Gonon, A.; Pernow, J.; Stangl, V.; Aspenström, P.; Kappert, K.; Östman, A. Cell. Signal. 2011, 23, 820. (37) He, R.-J.; Yu, Z.-H.; Zhang, R.-Y.; Zhang, Z.-Y. Acta Pharmacol. Sin. 2014, 35, 1227. 10 (38) Barr, A. J.; Ugochukwu, E.; Lee, W. H.; King, O. N. F.; Filippakopoulos, P.; Alfano, I.; Savitsky, P.; Burgess-Brown, N. A.; Muller, S.; Knapp, S.; Müller, S.; Knapp, S. Cell 2009, 136, 352. (39) Low, J.-L.; Chai, C. L. L.; Yao, S. Q. Antioxid. Redox Signal. 2014, 20, 2225. 11 Figure 1.1 Classification and general structural features of the PTP superfamily Reprinted with permission from Zhang, Z.-Y. Acc. Chem. Res. 2003, 36, 385. Copyright (2003) American Chemical Society. 12 a b Figure 1.2 The conserved catalytic mechanism of PTPs. a. A ribbon diagram of the Yersinia PTP tertiary structure showing the location of active site P-loop, WPD loop, and the Q-loop. Reprinted with permission from Zhang, Z.-Y. Acc. Chem. Res. 2003, 36, 385. Copyright (2003) American Chemical Society. b. The catalytic mechanism and transition states of the dephosphorylation reaction, demonstrated with Yersinia PTP. Reprinted with permission from Zhang, Z.-Y. Acc. Chem. Res. 2003, 36, 385. Copyright (2003) American Chemical Society. 13 Figure 1.3 The reaction pathways of PTPs under oxidation and reduction. CHAPTER 2 RATIONAL DESIGN OF SHP2-SPECIFIC FLUOROGENIC SUBSTRATES AS MECHANISM-BASED PROBES 2.1 Introduction The survival and proliferation of cancer cells often result from the aberration of critical cell signaling pathways including ERK-MAP, PI3K-AKT, and NF-κB, all of which are tightly regulated by the combined actions of protein tyrosine phosphatases (PTPs) and protein tyrosine kinases (PTKs). Since many PTKs are identified as oncoproteins, it was generally believed that most PTPs act as tumor suppressors and prevent the propagation of pro-survival signals. However, a prominent exception is SHP2, which was identified as the first oncoprotein in the PTP superfamily. Mutations in SHP2 are believed to cause Noonan syndrome and LEOPARD syndrome,1,2 and elevated SHP2 activities are found in different types of malignancies.3 In addition, SHP2 is a key mediator in the T cell receptor (TCR) signaling pathway and could be a potential target for tumor immunotherapy.4 Based on these observations, SHP2 has garnered significant attention in cancer research. Most biochemical studies identify an upregulation or downregulation of SHP2 by monitoring the change in expression level, using either RT-PCR or immunoblotting with corresponding antibodies. However, directly correlating the expression level to the enzymatic activity level can be misleading. Due to the unique microenvironment in the 15 active site, the catalytic cysteine residue of SHP2 (like all PTPs) has a perturbed pKa value (~4.5) which makes it prone to oxidation. Researchers have demonstrated that the activity of SHP2 can be largely attenuated by reactive oxygen species (ROS), which are produced in response to a variety of stimuli, including growth factors, cytokine, and UV light.5 Traditional biochemical methods, such as RT-PCR or immunoblotting, cannot differentiate the inactive oxidized enzymes from the active ones, and therefore can lead to misinterpretation of the real impact of transcriptional changes on enzymatic activities in cell signaling pathways. Chemical probes, on the contrary, can provide an accurate readout of the actual enzymatic activities. Nevertheless, commercially available PTP chemical probes, such as DiFMUP and pNPP, are nonspecific towards most PTPs as well as other protein phosphatases. In addition, due to the high structure conservation in PTP catalytic site, it is very challenging to develop small chemical probes with good specificity towards certain PTPs over the others. To overcome the limitations of existing chemical probes, pCAP, a phosphotyrosine mimic that can be hydrolyzed by a wide range of PTPs to give a large increase in fluorescent signal, was previously developed in our lab.6 The pCAP amino acid can be easily incorporated into short peptides using standard solid phase peptide synthesis method. Thus, in a pCAP-based probe, the pCAP moiety can provide a sensitive and convenient readout of enzyme activity, while the flanking peptide residues would interact with residues surrounding the enzyme active sites, which are less conserved among different PTPs, to provide selectivity. The strategy has been validated previously with pCAP-based probes targeting on TCPTP (Table 2.1).7 By changing the peptide sequences, we were able to improve the 16 probe from a bad substrate to a moderate one, and eventually to a good substrate for TCPTP. Moreover, a CD45-selective pCAP probe was employed by Szodoray et al to reveal an upregulation of CD45 activity in systemic lupus erythematosus (SLE) patients compared to healthy controls.8 Using an antibody, the expression level of CD45 in BND cells from different patient groups was determined to be similar. However, the CD45 phosphatase activities in both active or inactive SLE patients were clearly higher than healthy patients, as determined by a CD45-selective pCAP-based probe developed by our lab (Figure 2.1).9 To help reveal the activity of SHP2 in key oncological signaling pathways, we designed a SHP2-selective mechanism-based fluorescent probe using pCAP and a selective peptide sequence, as illustrated in Figure 2.2a. 2.2 Results and discussion To determine a selective sequence for SHP2, we took a rational approach and attempted to identify a sequence from a series of potential biological substrates of SHP2 whose reactivities were previously reported by Barr et al.10 We extracted a candidate sequence from a potential biological substrate of SHP2, tyrosine kinase TIE2 (DPTIpY816-PVLD),10 and synthesized a pCAP-based probe (DPTI-pCAP peptide) as illustrated in Figure 2.2b. The probe was tested with SHP2 and several other PTPs that can potentially cross-react with this peptide sequence, according to Barr et al.10 It exhibits a high micromolar dissociation constant (KM = 0.7 ± 0.2 mM) towards SHP2, which falls in a similar range compared to the commercially available general PTP substrate pNPP.11 The kcat of the probe is relatively low, but could potentially be further optimized by tuning the 17 sequence. Data showing the hydrolysis of the DPTI-pCAP peptide by SHP2 are shown in Figure 2.3a. A comparison of initial velocities (V0) with a panel of PTPs is shown in Figure 2.3b. This sequence showed more than 8-fold selectivity towards SHP2 over PTPmu, HePTP, LYP, and PTP-MEG1, but the peptide was hydrolyzed equally well by SHP2 and TCPTP, raising some concerns regarding cross-reactivity. In addition, the DPTI-pCAP peptide proved challenging to synthesize. During the solid phase peptide synthesis, pCAP is coupled in an ethyl ether protected form to prevent side reaction on the phosphate hydroxyls. The two ethyl groups are cleaved using trimethylsilyl iodide (TMSI) right after the last coupling cycle when the peptide is still bound to the resin. However, the DPTI-Et2pCAP-PVLD showed an evident deficiency during the TMSI deprotection. Aiming at further enhancing the substrate reactivity and potentially improving the probe selectivity, we summarized a full list of natural substrates that have decent reactivities with SHP2 from reference,10 as shown in Figure 2.4a. By using the positional amino acid abundance of SHP2 substrates, we assembled a list of “optimized” variants to the initial probe (P1 to P6, Figure 2.4b). All of the variants synthesized exhibited enhanced reactivity compared to the initial probe, with P2 exhibiting a significantly higher reactivity compared to the other peptides (Figure 2.5a). More importantly, P2 remains selective towards SHP2 over LYP, HePTP, PTP-MEG1, and PTPmu, although its reactivity towards TC-PTP also increased to a similar level as SHP2, as shown in Figure 2.5b. In certain subsets of breast cancer cell lines, such as BT-483, MDA-MB-134, MDA-MB-157, and MDA-MB-175 cells, TC-PTP, which serves as tumor suppressor, was found largely deficient, while SHP2 is overexpressed and promotes tumor 18 maintenance and invasion.12,13 With the good reactivity and decent selectivity, P2 will be able to serve as a SHP2-specific mechanism-based probe to measure SHP2 activities in such cell lines to help establish a better understanding of the physiological roles of SHP2 during cancer progression. We further explored the possibility of applying the probe in a cellular environment by incorporating a cell-penetrating tag into the sequence. Cell-penetrating peptides are known to facilitate the internalization of peptides; the most common ones are short arginine peptides.14 Four variants of P2 probe, P92_1 through P91_4 (Figure 2.6a), with R5 and R8 poly-arginine tags at either N-terminus or C-terminus, were designed and synthesized. Enzyme analysis indicated that the cell-permeable variants of SHP2 probes were still reactive towards SHP2, although the reactivity dropped almost 10-fold (Figure 2.6c). Due to a slightly basic catalytic pocket, low reactivity of basic chemical probes is not uncommon for PTP targeting probes. With the significantly reduced potency, it is not surprising that none of these cell-permeable probes gave observable fluorescent signal when tested with PC9 cells, a lung cancer cell line that is known to express active SHP2.15 2.3 Conclusion A series of SHP2-specific fluorescent substrates were designed and synthesized based on a rational approach. The potency and selectivity of these probes were validated with a panel of PTPs. The results indicated that peptide-based, specific SHP2 substrates can be designed and optimized using the sequence information from potential biological substrates. Using this approach, the P2 probe achieved a good potency towards SHP2 and decent selectivity over a list of PTPs. We also demonstrated that directly appending a 19 highly basic, poly-arginine cell-penetrating peptide to the probe greatly decreases the probe potency. Based on these results, we conclude that a more potent parent sequence with better kinetic parameters, particularly kcat, is needed to overcome the reduction in activity resulting from addition of the basic cell-penetrating tag. However, due to the limited substrate information from literature, rationally designing a cell-permeable SHP2specific mechanism-based probe remains challenging. 2.4 Experimental section All reagents and organic solvents were obtained from commercial sources and used without further purification. All NMR spectra were recorded on a Mercury 400 MHz spectrometer and were referenced to residual solvent peaks or TMS (d 0.00 ppm). Column chromatography was performed on EMD silica gel (60-200 mesh) or EMD alumina (80-200 mesh). The SHP2 enzyme was prepared using pET21a-SHP2 plasmid from Zhong-Yin Zhang’s lab.16 Enzyme kinetic assays were run at rt on a Molecular Devices Spectramax M5 multimode plate reader with excitation and emission at 360 and 455 nm, respectively. 2.4.1 Synthesis of pCAP The synthesis of pCAP follows an optimized protocol in 4 steps, starting from intermediate 3, as illustrated in Figure 2.7. 20 2.4.1.1 Synthesis of 6-methyl 1-(2,2,2-trichloroethyl) 2-((((9H -fluoren-9-yl) -methoxy) -carbonyl) -amino)-4 -oxohexanedioate (4), Figure 2.7, step (iii) A solution of 3-((((9H-fluoren-9-yl) -methoxy) -carbonyl) -amino)-4-oxo-4(2,2,2-trichloroethoxy) -butanoic acid (487 mg, 1 mmol), Meldrum’s acid (216.3 mg, 1.5 mmol), and 4-dimethylaminopyridine (244.3 mg, 2 mmol) in 5 mL of dichloromethane was cooled down to 0°C in an ice-salt bath. To this reaction mixture, a pre-chilled solution of 1-Ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochloride (287.7 mg, 1.5 mmol, in 2 mL dichloromethane) was added dropwise under N2 while stirring. The reaction was allowed to stir at 0°C for 4 h, then gradually warmed up to rt and stirred overnight. The reaction mixture was then quenched by 5% KHSO4 aqueous solution and diluted with dichloromethane, washed with water and brine, and dried over Na2SO4. The resulting crude mixture was concentrated and re-dissolved in 15 mL of MeOH/benzene (1:4) and refluxed for 4 h. Evaporation of solvent and column purification gave final product 4 as colorless liquid, yield 31%. 1H NMR (CDCl3, TMS, 400 MHz) d 7.76 (d, J=7.2, 2H), 7.60 (d, J=7.2, 2H), 7.41 (t, J=7.2, 2H), 7.32 (t, J= 7.2, 2H), 5.82 (d, J= 8.8, 1H, NH), 4.81 (d, J=12, 1H), 4.77 (m, 1H), 4.73 (d, J= 12, 1H), 4.42- 4.46 (A part of an ABX system, JAB = 2.8, JAX=7.6, 1H), 4.33- 4.37 (B part of an ABX system, JAB = 2.8, JBX= 7.6, 1H), 4.23 (t, J = 7.2, 1H), 3.74 (s, 3H), 3.51 (s, 2H), 3.39-3.45 (A part of an ABX system, JAB= 18.8, JAX= 4, 1H), 3.19- 3.25 (B part of an ABX system, JAB= 18.8, J BX= 4, 1H). 21 2.4.1.2 Synthesis of 2,2,2-trichloroethyl 2-((((9H-fluoren-9-yl) -methoxy) -carbonyl) -amino)-3-(7-hydroxy-2-oxo-4a,8a -dihydro-2H-chromen-4-yl) propanoate (5), Figure 7, step (iv) To 6-methyl 1-(2,2,2-trichloroethyl)2-((((9H-fluoren-9-yl) -methoxy) -carbonyl) amino)-4-oxohexanedioate (4, 148 mg, 0.28 mmol) and resorcinol (46.64 mg, 0.42 mmol) was added methane sulfonic acid dropwise under N2 at 0°C. The reaction was allowed to proceed at 4°C for 6 h and monitored by TLC until completion. The resulting mixture was extracted with ethyl acetate and washed with cold water and brine, purified through silica column to give the final product as white solid, yield 59%. 1H NMR (CDCl3, TMS, 400MHz) d 8.86 (s, 1H), 7.76 (d, J=7.2, 2H), 7.63 (d, J= 8.4, 1H), 7.55 (d, J=7.2, 2H), 7.40 (t, J=7.2, 2H), 7.30 (t, J= 7.2, 2H), 6.88 (d, J= 2.0, 1H), 6.86 (d, J= 1.6, 1H), 6.16 (s, 1H), 5.97 (d, J= 8.4, 1H), 4.84 (d, J=12, 1H), 4.83 (m, 1H), 4.74 (d, J= 12, 1H), 4.39 (d, J= 7.2, 2H), 4.19 (t, J = 6.8, 1H), 3.40- 3.45 (A part of an ABX system, JAB= 17.6, JAX= 5.2, 1H), 3.16- 3.21 (B part of an ABX system, JAB= 17.6, J BX= 5.2, 1H). 2.4.1.3 Synthesis of 2,2,2-trichloroethyl 2-((((9H-fluoren-9-yl) -methoxy) -carbonyl) -amino)-3-(7-((diethoxyphosphoryl) oxy)-2-oxo-4a,8a-dihydro-2H-chromen-4-yl) propanoate (6), , Figure 7, step (v) To 2-((((9H-fluoren-9-yl) -methoxy) -carbonyl) -amino)-3-(7-hydroxy-2-oxo4a,8a-dihydro-2H-chromen-4-yl) propanoic acid (5, 605 mg, 1 mmol) and DIPEA (323 mg, 2.5mmol) in 20 mL chloroform was added diethyl phosphorochloridate (207 mg, 1.2 mmol) dropwise at rt. The reaction was allowed to stir overnight and TLC monitored 22 until completion. The reaction mixture was concentrated under reduced pressure and quenched with 5% aqueous KHSO4 and then washed with saturated NaHCO3 solution, H2O, and brine, extracted with ethyl acetate and purified by silica column chromatography to give final product 6 as a colorless liquid, yield 35%. 1H NMR (CDCl3, 400 MHz) d 7.75 (d, J= 7.6, 3H), 7.55 (d, J=7.6, 2H), 7.39 (t, J=7.6, 2H), 7.30 (t, J=7.6, 2H), 7.23 (m, 2H), 6.27 (s, 1H), 5.54 (d, J=7.2, 1H), 4.84 (d, J=12.4, 1H), 4.82 (m, 1H), 4.71 (d, J= 12.4, 1H), 4.41 (d, J= 6.8, 2H), 4.25 (m, 5H), 3.36- 3.41 (A part of an ABX system, JAB= 14.8, JAX= 5.6, 1H), 3.15- 3.20 (B part of an ABX system, JAB= 14.8, J BX= 5.6, 1H), 1.37 (t, J=6.8, 6H). 2.4.1.4 Synthesis of (S)2-((((9H-fluoren-9-yl) methoxy) carbonyl) amino)-3-(7-((diethoxyphosphoryl) oxy)-2-oxo-4a,8a-dihydro -2H-chromen-4-yl) propanoic acid (Et2pCAP, 7), Figure 7, step (vi) Compound 6 (260.8 mg, 0.35 mmol) and zinc dust (138.5 mg, 2.12 mmol) were placed in a round bottom flask, and 50% acetic acid in THF was added dropwise at rt. The reaction was allowed to stir overnight and TLC monitored until completion. The resulting mixture was filtered and the filtrate poured into water and extracted with diethyl ether (20 mL x 3). The combined extracts were washed with H2O and brine, dried with Na2SO4, and evaporated under reduced pressure to remove the excess solvent. Further purification by silica column chromatography gave the final product as white solid, yield 38%. 1H NMR (CDCl3, TMS, 400MHz) d 7.81 (d, J= 8.0, 1H), 7.74 (d, J=7.6, 2H), 7.55 (m, 2H), 7.37 (m, 2H), 7.30 (m, 2H), 7.18 (m, 2H), 6.27 (s, 1H), 5.76 (d, J=7.6, 1H), 4.68 (q, J=6.8, 1H), 4.38 (m, 2H), 4.24 (m, 5H), 3.26 (m, 2H), 1.37 (t, J=6.8, 6H). 23 2.4.2 Solid phase peptide synthesis (SPPS) of pCAP peptides Peptides were synthesized on an AAPTech Focus XC automated peptide synthesizer using TentaGel S Rink Amide resin (0.22 mmol/g). The Fmoc group was cleaved with 20% piperidine for 5 min three times, and the typical coupling reaction was carried out using 3 equiv of Fmoc-amino acids, 3 equiv of DICI, 5 equiv of HOBT, in anhydrous DMF and was allowed to react at rt with shaking for 90 min. For the pCAP peptides, 2 equiv of Fmoc-Et2pCAP (7) was coupled during the SPPS. The ethyl ether groups were removed on resin at the end of the synthesis with 1M of TMSI solution in anhydrous CHCl3 under Ar for 1.5 h at rt. 2.4.3 Peptide cleavage and ether precipitation The synthesized peptides were cleaved from the resin and deprotected using a cleavage cocktail containing 88% TFA (v/v), 5% phenol (v/v), 5% water (v/v), and 2% triisopropylsilane (v/v) for 6 h at rt. The cleavage solution was slowly precipitated in cold ether (10 times the volume of cleavage solution) and incubated in the freezer overnight. The precipitated peptides were collected through centrifugation and washed with 30 mL of cold ether twice, air-dried, and purified on reverse phase HPLC. 2.4.4 Protein expression and purification An aliquot of E.coli BL21(DE3) competent cells was transformed with 1 μL of plasmid solution. The successful transformants were selected by plating on an ampicillin agar plate, and subsequently inoculated into 4 x 10 mL overnight starter LB culture. The starter culture was then inoculated into 4 x 1 L of LB medium in baffled flasks and cultured at 30ºC until OD600 reached 1.0, then IPTG was added to 1 mM final 24 concentration to induce expression at 18ºC. The cell culture flasks were kept shaking at 18ºC overnight and harvested by centrifugation. The cell pellet was re-suspended in 80 mL of lysis buffer, added with phenylmethylsulfonyl fluoride (final conc. 0.2 M), BME (final conc. 10 mM), and lysozyme (final conc. 0.6 mg/mL), incubated on ice for 20 min, and lysed by sonication for 2 x 1.5-min cycles. Triton X-100 1% solution (10 µL/mL) was added in between the two cycles and incubated for 10 min. The cell lysate was then centrifuged at 14000 rpm for 20 min. After centrifugation, the supernatant was passed through a 0.45 µm filter and incubated with equilibrated Ni-NTA resin (Qiagen, ID30210) in cold room for 45 min, flowed through a column, washed by His-bind buffer (50 mL) and His-wash buffer (50 mL), and finally eluted with His elution buffer (3 x 10 mL) following manufacturer’s protocol. When necessary, the eluted product was dialyzed against PBS buffer and digested by thrombin protease, followed by subsequent Ni-NTA and benzamidine column to remove the thrombin and further increase purity. The product was then concentrated and loaded onto a S75 gel filtration column. The fractions were analyzed by SDS-PAGE electrophoresis, and the desired portion was combined and added with DTT before aliquoting into 200 µL/tube. The aliquots were flash frozen with liquid N2 and stored at -80ºC. 2.4.5 Enzyme kinetic assays All PTP enzymes were pre-activated with 100 mM TCEP for 30 min on ice prior to use. The kinetic assay was carried out using a 96-well black bottom plate, with an assay buffer containing 50 mM Tris, 100 mM NaCl, 0.01% Brij-35, and 2 mM EDTA, 25 pH 7.4. After the addition of substrate, the plate was immediately placed in the plate reader and the fluorescence was monitored for 60 min at 30 s intervals. The rate was calculated within the linear range with data obtained no later than 5 min. DiFMUP (5 µM) was used as a positive control when necessary. 2.5 References (1) Noda, S.; Takahashi, A.; Hayashi, T.; Tanuma, S. I.; Hatakeyama, M. Biochem. Biophys. Res. Commun. 2016, 469, 1133. (2) Tartaglia, M.; Mehler, E. L.; Goldberg, R.; Zampino, G.; Brunner, H. G.; Kremer, H.; van der Burgt, I.; Crosby, H.; Ion, A.; Jeffery, S.; Kalidas, K.; Patton, M.; Kucherlapati, R. S.; Gelb, B. D. Nat. Genet. 2001, 29, 465. (3) Ran, H.; Tsutsumi, R.; Araki, T.; Neel, B. G. Cancer Cell 2016, 30, 194. (4) Lorenz, U. Immunol Rev. 2009, 228, 342. (5) Chen, C. Y.; Willard, D.; Rudolph, J. Biochemistry 2009, 48, 1399. (6) Mitra, S.; Barrios, A. M. Bioorg. Med. Chem. Lett. 2005, 15, 5142. (7) Mitra, S.; Barrios, A. M. ChemBioChem 2008, 9, 1216. (8) Szodoray, P.; Stanford, S. M.; Molberg, Ø.; Munthe, L. A.; Bottini, N.; Nakken, B. J. Allergy Clin. Immunol. 2016, 138, 839. (9) Stanford, S. M.; Panchal, R. G.; Walker, L. M.; Wu, D. J.; Falk, M. D.; Mitra, S.; Damle, S. S.; Ruble, D.; Kaltcheva, T.; Zhang, S.; Zhang, Z.-Y.; Bavari, S.; Barrios, A. M.; Bottini, N. Proc. Natl. Acad. Sci. U. S. A. 2012, 109, 13972. (10) Barr, A. J.; Ugochukwu, E.; Lee, W. H.; King, O. N. F.; Filippakopoulos, P.; Alfano, I.; Savitsky, P.; Burgess-Brown, N. A.; Muller, S.; Knapp, S.; Müller, S.; Knapp, S. Cell 2009, 136, 352. (11) Sarmiento, M.; Zhao, Y.; Gordon, S. J.; Zhang, Z.-Y. J. Biol. Chem. 1998, 273, 26368. (12) Aceto, N.; Sausgruber, N.; Brinkhaus, H.; Gaidatzis, D.; Martiny-Baron, G.; Mazzarol, G.; Confalonieri, S.; Quarto, M.; Hu, G.; Balwierz, P. J.; Pachkov, M.; Elledge, S. J.; Van Nimwegen, E.; Stadler, M. B.; Bentires-Alj, M. Nat. Med. 2012, 18, 529. (13) Shields, B. J.; Wiede, F.; Gurzov, E. N.; Wee, K.; Hauser, C.; Zhu, H.-J.; Molloy, 26 T. J.; O’Toole, S. A.; Daly, R. J.; Sutherland, R. L.; Mitchell, C. A.; McLean, C. A.; Tiganis, T. Mol. Cell. Biol. 2013, 33, 557. (14) Futaki, S.; Suzuki, T.; Ohashi, W.; Yagami, T.; Tanaka, S.; Ueda, K.; Sugiura, Y. J. Biol. Chem. 2001, 276, 5836. (15) Nichols, R. J.; Haderk, F.; Stahlhut, C.; Schulze, C. J.; Hemmati, G.; Wildes, D.; Tzitzilonis, C.; Mordec, K.; Marquez, A.; Romero, J.; Hsieh, T.; Zaman, A.; Olivas, V.; McCoach, C.; Blakely, C. M.; Wang, Z.; Kiss, G.; Koltun, E. S.; Gill, A. L.; Singh, M.; Goldsmith, M. A.; Smith, J. A. M.; Bivona, T. G. Nat. Cell Biol. 2018, 20, 1064. (16) Zhang, X.; He, Y.; Liu, S.; Yu, Z.; Jiang, Z.; Yang, Z.; Chan, J.; Zhang, Z. J. Med. Chem. 2010, 53, 2482. (17) Crooks, G.; Hon, G.; Chandonia, J.; Brenner, S. Genome Res. 2004, 14, 1188. 27 Figure 2.1 The expression level and phosphatase activity of CD45 in BND cells from patients with active or inactive SLE, compared with healthy control.8 Figure reprinted from Journal of Allergy and Clinical Immunology, 138(3), Szodoray, P., Stanford, S. M., Molberg, Ø., Munthe, L. A., Bottini, N., & Nakken, B., T-helper signals restore B-cell receptor signaling in autoreactive anergic B cells by upregulating CD45 phosphatase activity, 839-851, Copyright (2016), with permission from Elsevier. 28 a. peptide H N O C peptide b. H Ac-DPTI N O C PVLD-amide O OH HO P O O O O OH HO P O O O Figure 2.2 The design of SHP2-specific mechanism-based probes. a. A general scheme of the probe design b. The DPTI-pCAP peptide Table 2.1 Three pCAP peptides with different sequences and reactivities towards TCPTP. NR: no reaction observed. Kcat/KM (1/sM x 10-4) Bad DPHR-pCAP-VWKR NR NR NR Mediocre VFDQ-pCAP-HESP 11.5 ± 0.5 0.28 ± 0.03 4.1 ± 0.5 Good FnGA-pCAP-QLEE 160 ± 30 0.27 ± 0.05 60 ± 10 Substrate Sequence kcat (1/s) KM (mM) 29 b. a. c. kcat (1/s) Km (mM) kcat/Km (1/sM) DPTI-pCAP-PVLD 3.1 ± 0.3 0.7 ± 0.2 5000 ± 2000 pNPP 17.3 ± 0.9 2.2 ± 0.4 7900 ± 1500 Figure 2.3 Kinetic characterizations of the DPTI-pCAP-PVLD peptide. a. Enzyme kinetics of DPTI-pCAP-PVLD peptide with SHP2. b. A comparison of initial reaction rate between SHP2, TCPTP, PTPmu, HePTP, LYP, and PTP-MEG1. c. A comparison on kinetic parameters between the DPTI probe and the commercially available general pNPP probe. 30 a. 2 3 4 T 1 -2 P 0 -3 D -1 -4 b. I pCAP P V L D # P1 P2 P3 P4 P5 P6 Sequence DDTI-pCAP-VVLD DDTI-pCAP-DVLD DDTI-pCAP-TVLD DSTI-pCAP-VVLD DSTI-pCAP-DVLD DSTI-pCAP-TVLD Figure 2.4 The design of optimized variants of the DPTI-pCAP-PVLD peptide. a. A summary of the frequency of amino acids found at each position of SHP2 substrates. Position 0 is the phosphorylated tyrosine. Sequence logo was generated using WebLogo, developed by Dr. Schneider.17 b. a list of substituted proline-free peptide candidates. 31 a b Figure 2.5 Kinetic characterizations of the optimized variants P1 to P6. a. Enzyme assay of the series of probes with SHP2. b. A comparison of enzyme reactivities of P1- P6 with a panel of PTPs. 32 a c DD TI R -pC A 5 -D DT P-D I-p VL R CA D 8 -D PDT I-p DVL D DD CA TIPD pC VL AP D DD D TIV L pC A P D- R DP TI- DVL 5 D pC AP -R -P 8 VL D b Figure 2.6. Kinetic characterizations of the cell-permeable variants of P2 probe. a. The design of P92_1 to P92_4 as cell-permeable variants of P2 probe. b. Enzyme assay of the cell permeable variants of P2. c. Comparing the reactivities of P2 with the cellpermeable variants of P2. 33 Figure 2.7 Synthesis of pCAP. Reagents and conditions: synthesis of 7 (pCAP) (i) CCl3CH2OH, EDCI, DMAP, 0 oC, 15 h, 90%. (ii) 50% TFA in CH2Cl2, rt, 4 h, 95%. (iiia) EDCI, Meldrum’s acid, DMAP, CH2Cl2, 0 oC, 16 h. (iiib) CH3OH/benzene (4:1), reflux, 12 h, 31%. (iv) Resorcinol, CH3SO3H, 0 oC, 6 h, 59%. (v) (EtO)2POCl, diisopropylethylamine, CHCl3, rt, 14 h, 35%. (vi) Zn dust (6 equiv) 50% acetic acid in THF, rt, 8 h, 38%. CHAPTER 3 SYNTHETIC PEPTIDE LIBRARY SCREENING FOR PTPS 3.1 Introduction Previous work in our lab has demonstrated peptide-based probes to be useful research tools in investigating the biological roles of PTPs.1,2 However, the biggest challenge in developing peptide-based probes is obtaining a selective peptide sequence. Many PTPs do not have a well-characterized list of physiological substrates. For PTPs that have been more closely investigated, such as TCPTP and CD45, a few important physiological substrates have been identified. Nevertheless, these substrates do not exhibit satisfying selectivity and often cross-react with other PTPs. Another challenge lies in the bioavailability of peptide substrates, as many of the physiological substrates contain multiple acidic residues which would reduce cell permeability, thus hindering their application in cellular experiments. Since it is challenging to rationally design a PTP probe with good selectivity and bioavailability (as highlighted in Chapter 2), we decide to solve the problem through a combinatorial peptide library screening. A screen with a randomized peptide library has the potential to identify sequences which provide better selectivity than natural substrates if sufficient chemical space is covered. The hit sequences can then be converted into selective PTP probes by replacing the pY position with functional, reactive molecules including substrates or inhibitors. 35 Phage display has been a popular biochemical method for generating large randomized libraries. However, the phage display libraries lack important posttranslational modifications, including tyrosine phosphorylation, which is critical for substrate recognition in the PTP catalytic site. In addition, phage display library peptides only consist of natural amino acids, which limits the chemical and structural diversity of the library. To address these challenges, we decided to use an inverse alanine scanning method to obtain positional preference profiles for multiple PTPs, from which we can assemble optimal sequences for each individual PTP to ensure adequate selectivity. In the inverse alanine scan approach, only one position at a time will be changed to different amino acids, while all the other positions are occupied by alanine. Solid phase peptide synthesis (SPPS) is a well-established technique and a diverse range of building blocks, including unnatural and phosphorylated amino acids, are commercially available and can be easily incorporated into the peptide sequence in high yield. Therefore, we assembled the inverse alanine scanning library using SPPS method, and the fluorescent pY-mimetic probe (pCAP) was incorporated to serve as a reporter of PTP activity. To improve the cell permeability of the peptide probes, we previously had success by adding a polyarginine tail to a CD45-selective probe to make it cell permeable towards Jurkat T-cells.1 However, when this same strategy was applied to peptide-based substrates of other PTPs such as PTPN22, the addition of the polyarginine tail reduced enzymatic reactivity significantly and had devastating effects on probe potency and selectivity. This phenomenon was also observed and discussed in Chapter 2, where addition of polyarginine largely disrupted the reactivity of the cell 36 permeable SHP2 probe. Previous experiments have shown that increasing the linker length between the polyarginine and peptide does not necessarily resolve this issue. Implementing a cleavable linker would potentially resolve the issue, but cleavage of a linker, either chemically or enzymatically, will be dependent on other factors in the cell and ultimately decrease the overall reliability of the probe. To resolve these challenges, we decided to incorporate a cell-penetrating peptide motif directly into the template of the library design. 3.2 Results and discussion To determine whether the change of one residue would introduce enough difference on the overall binding affinity to be observed, we compared an all-alanine peptide (C14-R7-AAAA-pCAP-AAAA) with one which contains a single amino acid change (C14-R7-AAAE-pCAP-AAAA). The initial velocities (V0) were measured and compared between a panel of PTPs at three substrate concentrations (50, 100, and 250 µM). The results are shown in Figure 3.1. CD45, VHR, TCPTP, SHP2, and HePTP showed a clearly higher V0 when the alanine is replaced with glutamate, while the remaining PTPs do not show any appreciable preference. Overall, these results indicated that the change of one amino acid can be sufficient to change the reactivity of the substrate. Another important part of the library design is to select a cell-penetrating motif to be incorporated in the peptide sequence template. Although many cell-penetrating peptides (CPP) can help internalize a cargo peptide, their efficiencies often vary depending on the cell line being tested. To select the optimum CPP for PC9 cells, a lung cancer cell line known to express active SHP2, we tested a series of CPPs from the 37 literature with different entry mechanisms as shown in Table 3.1. All of the listed CPPs were synthesized bearing a fluorescein tag as illustrated in Figure 3.2. The library of CPP-fluorescein peptides was incubated with PC9 cells for 3 h, and the internalization efficiencies were evaluated by measuring the fluorescent intensities of the cells using fluorescent microscopy. The relevant toxicities of these CPPs were also recorded by comparing changes in cell shape (Figure 3.3). The poly-arginine peptides gave the strongest signal and the signal intensity is positively related to the number of arginine. The attachment of a lipid tail to the poly arginine peptides further increased the signal intensity with the cost of dramatically impairing cell viability. Peptide 12 with the longest lipid chain displayed extremely intense signals, but caused severe damage to the cells and resulted in cell detachment (Figure 3.3). The HIV Tat peptide (peptide 5) did not affect cell viability, but showed less fluorescence intensity compared to the poly arginine peptides. The CAYHRLRRC sequence, reported by Nishimura et al3, gave very weak signals with both the truncated form (peptide 3) and the Cys-Ser substituted form (peptide 1 and 2). In summary, polyarginine sequences containing 5 to 8 arginine residues all proved to be effective in penetrating into the cells with low cytotoxicity. After the cell-penetrating motif was determined, the screening library was designed as illustrated in Figure 3.4. Eleven building blocks (b1-12, Figure 3.4 b) were screened at each position of a 9-mer peptide template to obtain a positional preference profile for each PTP of interest. The natural amino acids were grouped into 8 building blocks according to their chemical property similarities to reduce the library redundancy. To ensure an equal distribution of coupling in wells with more than one amino acid, the ratio of different amino acid residues was adjusted according to their 38 chemical reactivity, and these building blocks were used in 10-fold excess during the coupling reaction.4 To further expand the diversity of the library, an aromatic unnatural amino acid (naphthyl-alanine), an alkylated cysteine, and a mixture of phosphorylated serine/threonine/tyrosine were also added into the building blocks composition. The library of peptides was synthesized, purified, dissolved in DMSO, and transferred to a 96-well storage plate. The concentration of each peptide was measured based on the UV absorption at 320 nm and calculated using 4-methylumbelliferyl phosphate as the standard (ɛ320 = 3.3 ± 0.3 mM-1 cm-1, Figure 3.5a). The stock solutions were normalized to be 2 ± 0.2 mM (average = 1.99 mM, standard deviation = 0.115) before proceed to the enzyme screening. Stock solutions of DiFMUP (A12-D12: 5, 25, 50, 200 µM) and MUP (E12-H12: 20, 100, 200, 400 µM) were placed in last column of the library to serve as standard positive controls of the enzyme activity. The final concentrations of the stock solutions are listed in Figure 3.5b. The yield of each library member was calculated based on the theoretical yield from 3 µmol resin and the yields are listed in Figure 3.5c. The percentage yields are not very accurate because the isolation of resins in slurry created significant variations in the actual amount of resin being placed in each well. The normalized library was then screened at a 50 µM concentration with 11 enzymes: SHP2, HePTP, LYP, PTP-MEG1, PTP-e, LMWPTP, PTPµ, VHR, PTP1B, TCPTP, CD45, YopH. Since the substrate concentration in the assay is far less than the its Km, the initial velocity V0 is proportional to the corresponding kcat/Km of the substrate. Thus, V0 was used to evaluate the relative substrate reactivity (Table 3.2). For each enzyme, the V0 of one substrate was compared to other building blocks screened at the same position. The building block providing the highest reactivity in a 39 tested position was scored 100%, while other, less reactive building blocks were normalized accordingly (Table 3.3). These normalized data represent how much a certain building block is preferred at a given position for the tested enzyme. The preference for acidic residues at many positions is consistent with previous findings in the literature that PTPs prefer acidic residues at most sites due to the slightly basic catalytic pocket5. An interesting observation in the library screening data is that phosphorylated residues are generally preferred at all positions. This may be a result of the acidic character of phosphorylated residues, given the preference for acidic residues within the peptide. Furthermore, many native substrates of the PTPs contain multiple phosphorylated residues, such as the insulin receptor Y1150-Y1151.6 While this trend holds true in most cases, exceptions were found within the data set. For example, at the X4 position, phosphorylated residues are largely preferred, while substrates bearing Glu or Asp did not exhibit the same level of reactivity. Through an analysis of the library data, it can also be concluded that some positions are more sensitive to changes over others. One such example is the X6 position. When changed from acidic residues to nonpolar or aromatic residues, the reactivity dropped significantly. Another example is X3, which clearly prefers phosphorylated or acidic residues over basic residues. The X8 position is inert to most changes except when changed to the phosphorylated building blocks, which then showed an increase in activity. These observations provided further validation of the inverse alanine library for two reasons. First, enzymes showed distinctive preferences at each given position, such as the pS/pT at X6, which is highly favored by SHP2 and VHR but disfavored by CD45 or HePTP (Figure 3.6). The absence of a universal bias in certain residue-position 40 combination across screens ruled out major normalization or purity issues in the library. Secondly, the general preference for acidic residues is in agreement with conclusions from previous library screens found in the literature,7 which indicate that enzymatic selectivity is truly revealed by the data processing approach we used in this library. The ultimate goal of this library screen is to find selective sequences based on the inverse alanine scan data. The challenge is to define selectivity and identify key differences in substrate preferences. From our observations, it is common for one PTP to share similar preferences with other PTPs. For example, LYP follows all the general trends observed with other PTPs and has little unique preference of its own. There are a few PTPs that show unique differences when compared to other PTPs. LMWPTP demonstrated a clear preference for basic residues at X1/X6 position and aromatic residues at the X8 position, which is not observed with other PTPs. For PTPµ, we observed increased activity from aromatic residues at X2 and basic residues at X6, which usually prefer phosphorylated and acidic amino acids. Therefore, we summarized these observations and assembled a pool of lead sequence candidates in Table 3.4. To validate these lead candidates from the inverse alanine screen, we synthesized two of the “optimal” sequences for LMWPTP and PTPµ to test with the corresponding enzyme. However, for the building blocks containing more than one amino acid, it was necessary to identify which residue contributed the most to the enhanced reactivity. To assist in picking a sequence, several known natural substrates were aligned with our screen leads to help determine the lead sequences to be synthesized (Table 3.5). We also assembled a “bad probe” consisting of the least preferred amino acids at each position, AAKA-pCAP-P-(NaphAla)-LH-amide, to compare with the “good” probes and further demonstrate whether the screen data can 41 indeed guide the design of substrates for different PTPs. All three sequences were synthesized bearing a polyarginine tail (Ac-R5-) at the N-terminus. The results of these initial sequence optimization efforts are shown in Figure 3.6. The optimized probes did exhibit a better reactivity compared to the general bad probe, although the level of enhancement varies. These data indicate that it is possible to optimize future peptide substrates with the information from the inverse alanine screen and generate more potent and more selective fluorescent probes or inhibitors. It is worth noting that since many PTPs share similar preference within the 8 positions we scanned, it remains challenging to selectively target one PTP over the others using the catalytic site targeting approach. However, with the preference profile, it is possible to selectively optimize a probe to be more reactive toward one PTP over the other. This can be seen by comparing the activities of the “bad probe” towards both LMWPTP and PTPµ. Although being a generally disfavored probe, LMWPTP has a higher preference score for K at X3 position (37%, vs PTPµ’s 0%) and for Naphthyl Alanine at X6 position (20%, vs PTPµ’s 0%) comparing to PTPµ. This resulted in a higher reactivity of the bad probe towards LMWPTP and a smaller activity enhancement when compared to the “good” LMWPTP probe. 3.3 Conclusion A synthetic peptide library was designed and synthesized to identify the substrate sequence preference of PTPs at 8 positions. A panel of 11 PTPs was screened using the library and their positional preference profiles were obtained and analyzed. A peptide-based fluorescent substrate was designed and synthesized based on the positional preference profile of PTPµ and LMWPTP. 42 3.4 Experiments and methods All reagents and organic solvents were obtained from commercial sources and used without further purification. pCAP was synthesized in the lab following the procedures described in Chapter 2. All NMR spectra were recorded on a Mercury 400 MHz spectrometer and were referenced to residual solvent peaks or TMS (d 0.00 ppm). Column chromatography was performed on EMD silica gel (60-200 mesh) or EMD alumina (80-200 mesh). The TCPTP, CD45, and VHR enzymes were purchased from commercial sources. Other PTP catalytic domains were prepared using plasmids from Dr. Zhong-Yin Zhang’s lab (SHP2224-528,8 PTPµ867-1452, PTPep,9 PTP-MEG1637-926) and Dr. Nunzio Bottini’s lab (LYP,10 LMWPTP,11 HePTP44-339). Enzyme kinetic assays were tested following the same procedures described in Chapter 2. Mass spectrometry was recorded on a Micromass Q-ToF micro mass spectrometer. 3.4.1 Solid phase peptide synthesis The peptide library is synthesized on TentaGel S Rink Amide resin (0.22 mmol/g) in 10 µmol scale using manual peptide synthesis filter tubes. The Fmoc group on resin and amino acids were cleaved with 20% piperidine for 10 min twice, and the typical coupling reaction was carried out using 3 equiv. of Fmoc-amino acids, 3 equiv. of DICI, and 5 equiv. of HOBT in anhydrous DMF for 90 min at rt with gentle shaking. For the pCAP peptides, 1.5 equiv. of Fmoc-Et2pCAP was coupled during the SPPS. For the difficult residues (phosphorylated residues and arginine in the CPP tag), 3 equiv. of Fmoc-amino acids, 2.95 equiv. of HBTU, 5 equiv. of HOBT, and 6 equiv. of DIPEA were used for a first round of coupling. After 1.5 hr, the resins were rinsed with DMF and added with another 3 equiv. of Fmoc-amino acids, 3 equiv. of DICI, and 5 equiv. of 43 HOBT for a second round of coupling. After the last residue was coupled, all the peptides were acetylated at the N-terminus using 3 equiv. acetic acid, 2.95 equiv. HCTU, and 6 equiv. DIPEA. All coupling steps were closely monitored by using the ninhydrin test. 3.4.2 Peptide library deprotection, cleavage, and ether precipitation Aliquots of each peptide (3 µmol of resin) were isolated and placed in a 96-well filter plate (AcroPrep, PN 5056) to proceed to the corresponding deprotection and cleavage steps. An approximately 3 µmol aliquot of each peptide resin was placed in a 96-well filter plate. The ethyl ether protecting groups in Et2pCAP were cleaved using 1M TMSI in CDCl3 under argon. After rinsed thoroughly with DCM and MeOH, 150µL of Reagent B (88% TFA, 5% phenol, 5% water, 2% Triisopropylsilane (v/v)) was added to each resin and the mix allowed to react overnight under argon, in the absence of light. Peptides containing Trp and His were reacted for 3 h instead of overnight to reduce side reactions. The cleavage solution was eluted into a 96-well collection plate, then slowly precipitated into 10-fold excess of cold ether and incubated in 4°C overnight to eliminate the protecting groups and excess scavengers. The precipitated peptides were collected through centrifugation and washed three times with 1 mL of fresh cold ether, then dried under vacuum. The dried precipitations were redissolved in ACN-H2O and dried with speed vacuum again to eliminate trace amounts of ether and TFA. The library was characterized using MALDI-MS before proceeding to the screening. 44 3.4.3 Solid phase peptide synthesis of fluorescein labeled CPP peptides library Fmoc-Lys(ivDde)-OH was initially coupled to 390 µmol TentaGel S Rink Amide resin (0.22 mmol/g) and followed by a b-Alanine as a spacer. The preloaded resins were then split and transferred into 13 manual synthesis vessels, and each peptide was synthesized by hand using the same protocol described above. At the end of coupling, the N-terminal amines were blocked by acetylation. The Lys(ivDde) was then deprotected by 2% hydrazine hydrate solution (2 mL, shake for 3 min and repeated 3 times) and rinsed extensively. The resins were then added with 3 equiv. 5(6)-FAM, 3 equiv. of DICI, and 5 equiv. HOBT, incubated on shaker for 20 h (avoiding light). The coupling was repeated again with fresh reagents, and the completion of reaction was monitored by ninhydrin test. The peptides were cleaved and purified following general protocols (avoid light). The fractions were analyzed by LC-MS and pure fractions were combined and lyophilized. The final yields of the peptides were listed in Table 3.5c. 3.4.4 Enzyme kinetic assays All PTP enzymes were pre-activated with 100 mM TCEP for 30 min on ice prior to use. The kinetic assay was carried out using a 96-well black bottom plate, with an assay buffer containing 50 mM Tris, 100 mM NaCl, 0.01% Brij-35, and 2 mM EDTA, pH7.4. After the addition of substrate, the plate was immediately placed on the plate reader for a continuous fluorescence reading for 60 min at 30 s intervals. The rate was calculated within the linear range with data obtained no later than 5 min. A 5 µM aliquot of DiFMUP was used as a positive control when necessary. 45 3.5 Reference (1) Stanford, S. M.; Panchal, R. G.; Walker, L. M.; Wu, D. J.; Falk, M. D.; Mitra, S.; Damle, S. S.; Ruble, D.; Kaltcheva, T.; Zhang, S.; Zhang, Z.-Y. Z.-Y.; Bavari, S.; Barrios, A. M.; Bottini, N. Proc. Natl. Acad. Sci. U. S. A. 2012, 109, 13972. (2) Szodoray, P.; Stanford, S. M.; Molberg, Ø.; Munthe, L. A.; Bottini, N.; Nakken, B. J. Allergy Clin. Immunol. 2016, 138, 839. (3) Nishimura, S.; Takahashi, S.; Kamikatahira, H.; Kuroki, Y.; Jaalouk, D. E.; O’Brien, S.; Koivunen, E.; Arap, W.; Pasqualini, R.; Nakayama, H.; Kuniyasu, A. J. Biol. Chem. 2008, 283, 11752. (4) Ostresh, J. M.; Winkle, J. H.; Hamashin, V. T.; Houghten, R. A. Biopolymers. 1994, 34, 1681. (5) Vetter, S. W.; Keng, Y.-F.; Lawrence, D. S.; Zhang, Z.-Y. J. Biol. Chem. 2000, 275, 2265. (6) Hashimoto, N.; Feener, E. P.; Zhang, W.-R.; Goldstein, B. J. J Biol Chem. 1992, 267, 13811. (7) Garaud, M.; Pei, D. J. Am. Chem. Soc. 2007, 129, 5366. (8) Yu, Z. H.; Xu, J.; Walls, C. D.; Chen, L.; Zhang, S.; Zhang, R.; Wu, L.; Wang, L.; Liu, S.; Zhang, Z. Y. J. Biol. Chem. 2013, 288, 10472. (9) Barr, A. J.; Ugochukwu, E.; Lee, W. H.; King, O. N. F.; Filippakopoulos, P.; Alfano, I.; Savitsky, P.; Burgess-Brown, N. A.; Muller, S.; Knapp, S.; Müller, S.; Knapp, S. Cell 2009, 136, 352. (10) Ahmed, V. F.; Bottini, N.; Barrios, A. M. ChemMedChem. 2014, 9, 296. (11) Stanford, S. M.; Aleshin, A. E.; Zhang, V.; Ardecky, R. J.; Hedrick, M. P.; Zou, J.; Ganji, S. R.; Bliss, M. R.; Yamamoto, F.; Bobkov, A. A.; Kiselar, J.; Liu, Y.; Cadwell, G. W.; Khare, S.; Yu, J.; Barquilla, A.; Chung, T. D. Y.; Mustelin, T.; Schenk, S.; Bankston, L. A.; Liddington, R. C.; Pinkerton, A. B.; Bottini, N. Nat. Chem. Biol. 2017, 13, 624. 46 a. b. c. d. Figure 3.1 A comparison of enzyme reactivities of the two test peptides. a. The summary of 10 PTPs reacting with 50 µM substrates. b. Michaelis-Menten curves of the two test peptide substrates with TCPTP. c.& d. Summary of 10 PTPs reacting with 100 µM & 250 µM substrates. 47 Table 3.1 A summary of the internalization efficiencies and toxicities of the CPP library. # Peptide sequence 1 Ac-HRLRRS-bAla-KFIuorCONH2 Ac-SAYHRLRRS-bAlaKFluor-CONH2 Ac-HRLRRC-bAla-KFluorCONH2 Ac-CAYHRLRRC-bAlaKFluor-CONH2 AcGRKKRRQRRRPPQbAla-KFluor-CONH2 Ac-R4 -bAla-KFluorCONH2 Ac-R5-bAla-KFluorCONH2 Ac-R6-bAla-KFluorCONH2 Ac-R7-bAla-KFluorCONH2 Ac-R8-bAla-KFluorCONH2 C12-R7-bAla-KFluorCONH2 C14-R7-bAla-KFluorCONH2 2 3 4 5 6 7 8 9 13 10 11 12 C16-R7-bAla-KFluorCONH2 Source Phage display library Possible mechanism Receptor independent, macropinocytosis HIV-Tat peptide Polyarginine peptides with varied length and lipophilicity Heperan sulfate proteoglycan dependent, receptor mediated endocytosis Fluorescence Cell viability level in cell No signal healthy Weak signal Healthy Weak signal Healthy Very weak signal Healthy Relatively bright signal Healthy Weak signal Healthy Bright signal Healthy Bright signal Healthy Bright signal Healthy Bright signal healthy Intense signal Cells look sick Intense signal Extremely intense signal Cells look very sick Cells severely damaged and detached 48 HN-Fmoc 1. PIP deprotection 1. PIP deprotection NH2-βAla-K(ivDde) 2. DICI coupling 2. DICI coupling Ac-HRLRRS-βAla-K(ivDde) Hydrazine hydrate Crude peptide TFA cleavage/ ether precipitation Ac-HRLRRS-βAla-K FITC DICI coupling, 5(6)-FAM 20 hrs, twcie, avoid light Ac-HRLRRS-βAla-K NH2 Microcleavage and analyzed by LC-MS Figure 3.2: The synthesis of fluorescein labeled CPP library. 8 4 Ac-CAYHRLRRC-bAla-KFluor-CONH2 Ac-R6-bAla-KFluor-CONH2 13 11 C14-R7-bAla-KFluor-CONH2 Ac-R8-bAla-KFluor-CONH2 Figure 3.3 Peptides 4, 8, 11, and 13 as examples of the fluorescent microscopy experiment in determining the internalization efficiency of the CPPs. 49 a Peptide template: Ac-R8-βA-X1X2X3X4-pCAP-X5X6X7X8-NH2 b1 b2 b3 b4 b5 b6 b7 b8 b9 b10 b11 b12 X1 Ac-R8-βA-X1AAA-pCAP-AAAA-NH2 X2 Ac-R8-βA-AX2AA-pCAP-AAAA-NH2 X3 X4 ... X5 X6 X7 Ac-R8-βA-AAAA-pCAP-AAAX8-NH2 X8 Hydrogen bonding 8 7 6 1 0 5 4 3 2 -1 -2 -3 1 0 Lypophilicity 2 H-bond acceptors H-bond donors LogP Number of H-donor/acceptors b 0 50 100 150 200 250 300 Molecular weight -4 0 50 100 150 200 250 300 Molecular weight Figure 3.4 The design of the reverse alanine screen library. a. A template of the positional scanned peptide library. b. Composition of the building blocks B1~12 and the distribution of their chemical properties. 50 a b. c. A B C D E F G H 1 2.11 2.12 2.07 1.86 2.07 2.10 2.02 2.16 2 2.03 2.06 2.1 1.8 1.89 1.92 2.05 1.81 A B C D E F G H 1 46% 39% 53% 35% 35% 28% 27% 30% 2 40% 39% 39% 44% 25% 19% 12% 23% 3 1.87 2.2 1.85 1.94 1.97 2.14 2.12 2.03 3 37% 58% 34% 34% 45% 25% 35% 36% 4 1.88 1.92 2.15 1.90 1.99 1.85 1.96 2.13 4 32% 29% 48% 38% 37% 37% 15% 27% 5 2.07 1.88 1.93 1.98 1.92 2.18 2.13 2.02 5 42% 40% 38% 32% 31% 29% 12% 26% 6 1.98 1.83 1.97 1.94 1.95 2.13 1.92 2.03 7 2.02 1.82 1.80 1.83 2.02 2.14 1.93 1.83 6 54% 78% 60% 64% 59% 32% 27% 27% 8 1.94 1.87 2.09 1.84 2.18 2.07 1.76 1.99 7 42% 41% 33% 44% 26% 21% 14% 11% 9 2.17 2.08 1.99 1.91 2.04 1.97 1.96 1.93 8 47% 45% 32% 29% 25% 28% 8% 33% 10 1.93 2.10 1.83 2.15 1.98 2.17 2.19 2.08 9 22% 71% 62% 74% 29% 24% 21% 18% 11 2.19 1.87 1.92 1.81 2.03 2.05 2.01 1.89 10 31% 58% 44% 66% 51% 10% 30% 9% 11 26% 41% 51% 29% 43% 34% 19% 21% Figure 3.5 Peptide library normalization and yields a. Standard curve of 4methylumbelliferyl phosphate, measured in 50% acetonitrile/H2O. b. Final concentration of each peptide in the library (mM). c. Percentage yields calculated from 3 µmol resin loading equivalents. 51 100% 80% 60% 40% 20% 0% Sh H p2 eP TP VH TC R PT P Yo pH LY M P EG PT 1 LM Pe W p PT PT P P C µ D4 5 Relative Activity Score pS/pT, at X4 position Figure 3.6 Enzyme preferences for pS/pT at X4 position. 52 Table 3.2 Library screen data V0, color coded. Blue: high reactivity. Red: low reactivity. Position Building Blocks SHP2 HePTP D/E 0.085 N/Q LVnI R/K X1 PTPe LMW PTP PTPµ CD45 VHR TCPTP YopH 0.009 0.014 0.068 -0.012 0.044 0.007 0.340 0.237 0.343 0.425 0.010 0.036 0.033 0.073 -0.010 0.007 0.006 0.186 0.327 0.299 0.353 0.025 0.014 0.004 0.031 -0.011 0.031 -0.008 0.107 0.206 0.192 0.202 -0.002 -0.012 0.029 0.027 0.003 0.062 -0.005 0.130 0.436 0.216 0.232 -0.030 0.019 0.005 0.016 -0.015 0.059 -0.043 0.244 0.370 0.284 0.336 F/Y/W -0.014 -0.013 0.000 0.022 -0.025 0.032 -0.036 0.098 0.336 0.286 0.221 H 0.062 0.003 0.003 0.032 0.003 0.002 0.012 0.155 0.296 0.282 0.304 P 0.105 -0.020 0.002 0.026 0.013 0.019 0.039 0.421 0.541 0.411 0.331 pS/pT 0.033 0.078 0.053 0.190 0.214 0.063 0.066 0.402 0.403 0.434 0.540 pY 0.059 0.044 0.069 0.112 0.022 0.042 0.035 0.175 0.286 0.273 0.404 -0.027 -0.018 -0.045 0.009 -0.022 0.034 -0.010 0.133 0.295 0.224 0.254 D/E 0.116 0.017 0.019 0.065 -0.041 0.026 0.006 0.651 0.296 0.258 0.365 N/Q 0.083 0.010 0.013 0.054 -0.007 0.040 0.013 0.780 0.372 0.235 0.250 0.090 -0.002 0.017 0.065 0.004 0.053 0.003 0.505 0.462 0.322 0.367 0.043 0.000 0.013 0.011 0.011 0.074 0.016 0.619 0.391 0.167 0.154 LVnI R/K S/T 0.073 -0.011 0.010 0.051 -0.016 0.060 0.004 0.717 0.366 0.236 0.303 F/Y/W 0.094 -0.023 0.011 0.054 0.021 0.051 0.041 0.633 0.430 0.249 0.237 H 0.070 0.007 0.011 0.053 0.005 0.028 0.023 0.426 0.284 0.226 0.225 P 0.095 0.094 0.194 0.028 0.023 -0.004 -0.016 0.821 0.644 0.340 0.395 pS/pT 0.128 0.049 0.011 0.070 -0.047 0.302 -0.053 1.144 0.461 0.479 0.599 pY 0.164 0.005 -0.018 0.048 -0.013 0.071 0.034 0.850 0.435 0.337 0.522 Naph-Ala 0.127 -0.018 0.017 0.056 0.003 0.033 0.000 0.653 0.589 0.334 0.282 D/E 0.195 0.024 0.056 0.104 -0.013 0.049 0.021 0.531 0.429 0.402 0.812 N/Q 0.099 0.014 0.036 0.053 0.027 0.038 0.009 0.226 0.403 0.266 0.280 LVnI 0.011 0.018 0.011 0.049 0.010 0.017 -0.016 0.089 0.242 0.224 0.247 -0.064 -0.019 -0.028 0.049 -0.030 0.058 -0.052 0.093 0.243 0.201 0.143 -0.029 -0.001 0.009 0.043 0.008 0.055 -0.042 0.200 0.296 0.263 0.366 R/K S/T X3 MEG1 S/T Naph-Ala X2 LYP F/Y/W -0.003 -0.011 -0.013 0.086 -0.025 0.051 -0.016 0.156 0.372 0.491 0.357 H 0.119 0.028 0.008 0.054 0.024 0.040 0.005 0.529 0.338 0.348 0.272 P 0.085 0.087 -0.049 0.021 -0.019 0.042 -0.048 0.928 0.363 0.344 0.296 pS/pT 0.134 0.035 -0.034 -0.008 -0.004 0.156 -0.072 0.864 0.572 0.391 0.505 pY 0.303 0.028 0.083 0.043 0.026 0.056 0.045 0.321 0.363 0.531 0.483 Naph-Ala 0.091 0.009 -0.006 0.067 -0.006 0.054 -0.016 0.171 0.326 0.532 0.321 53 Table 3.2. Continued Position Building Blocks D/E N/Q LVnI R/K S/T X4 F/Y/W H P pS/pT pY Naph-Ala D/E N/Q LVnI R/K S/T X5 F/Y/W H P pS/pT pY Naph-Ala D/E N/Q LVnI R/K S/T X6 F/Y/W H P pS/pT pY Naph-Ala SHP2 HePTP LYP MEG1 PTPe 0.126 0.021 0.019 0.080 0.006 0.075 0.010 0.012 0.044 0.090 0.006 0.016 0.071 0.016 0.109 LMW PTP PTPµ CD45 VHR TCPTP YopH 0.023 0.009 0.603 0.436 0.184 0.281 0.005 0.051 -0.008 0.701 0.389 0.177 0.239 0.053 0.004 0.007 0.005 0.607 0.491 0.203 0.191 0.001 0.021 0.067 0.034 -0.009 0.587 0.287 0.156 0.125 0.006 0.007 0.045 0.004 0.122 -0.005 1.042 0.555 0.335 0.309 0.078 0.011 0.014 0.044 0.034 0.004 0.002 0.771 0.495 0.211 0.152 0.097 0.013 0.010 0.068 -0.016 0.039 0.006 0.509 0.303 0.243 0.179 0.045 0.011 -0.015 -0.100 0.015 0.046 -0.025 0.926 0.297 0.120 0.215 0.272 0.083 0.148 0.046 -0.046 0.138 -0.020 1.063 0.507 0.355 0.383 0.289 0.019 0.064 0.051 -0.031 0.032 0.135 0.941 0.789 0.381 0.312 0.063 -0.062 -0.009 0.055 -0.001 0.010 -0.008 0.320 0.682 0.229 0.234 0.251 0.014 0.004 0.066 -0.011 0.000 -0.016 1.186 0.531 0.220 0.249 0.112 -0.033 -0.034 -0.022 0.007 0.030 -0.033 1.045 0.572 0.214 0.223 0.241 -0.010 -0.001 0.037 -0.003 0.003 0.012 0.848 0.584 0.201 0.131 0.054 -0.039 -0.020 0.034 -0.018 -0.013 0.014 1.171 0.595 0.126 0.063 0.097 -0.042 -0.024 0.009 0.021 0.043 -0.007 1.103 0.557 0.195 0.163 0.192 -0.089 -0.057 0.025 -0.017 -0.032 -0.011 0.948 1.094 0.088 0.114 0.188 0.004 -0.005 0.022 -0.004 0.016 0.032 1.152 0.514 0.249 0.179 -0.039 -0.052 -0.133 0.023 -0.393 -0.005 -0.029 1.116 0.684 0.054 0.072 0.163 0.022 -0.049 0.055 -0.027 -0.148 -0.067 0.952 0.626 0.201 0.239 0.286 0.018 0.004 0.021 0.000 -0.009 -0.025 1.382 0.637 0.193 0.154 0.220 -0.066 -0.078 0.032 0.100 -0.003 -0.008 0.481 0.965 0.135 0.083 0.211 -0.012 0.044 0.117 -0.043 0.055 0.004 0.370 0.475 0.406 0.527 0.053 0.012 0.025 0.026 -0.065 0.057 -0.033 0.196 0.679 0.245 0.270 -0.039 -0.003 0.007 0.027 0.018 0.015 -0.059 0.135 0.191 0.161 0.103 -0.063 -0.013 -0.044 -0.045 0.230 0.098 0.044 0.072 0.561 0.139 0.167 -0.047 -0.009 -0.017 0.033 0.035 0.046 -0.063 0.184 0.340 0.199 0.228 -0.057 -0.010 -0.022 0.062 -0.010 0.047 -0.050 0.050 0.386 0.133 0.233 0.047 0.026 0.011 0.033 -0.024 0.019 0.020 0.338 0.627 0.188 0.175 0.086 -0.020 -0.024 -0.016 -0.081 0.020 -0.032 1.466 0.554 0.303 0.214 0.339 0.016 0.007 0.060 -0.009 0.047 -0.011 0.402 1.362 0.493 0.598 0.161 0.059 0.001 0.042 -0.012 0.030 -0.007 0.096 0.487 0.136 0.199 -0.030 0.002 -0.077 -0.027 0.010 0.020 -0.020 0.039 0.412 0.071 0.067 54 Table 3.2 Continued Position Building Blocks D/E N/Q LVnI R/K S/T X7 F/Y/W H P pS/pT pY Naph-Ala D/E N/Q LVnI R/K S/T X8 F/Y/W H P pS/pT pY Naph-Ala SHP2 HePTP LYP MEG1 PTPe LMW PTP PTPµ CD45 VHR TCPTP YopH 0.029 -0.060 -0.074 0.009 -0.049 0.023 0.017 1.238 0.537 0.144 0.305 0.153 -0.018 0.008 0.034 -0.103 0.060 0.002 1.810 0.900 0.217 0.236 0.027 -0.030 -0.051 0.031 -0.017 -0.063 -0.018 0.130 0.623 0.201 0.377 0.016 -0.080 -0.023 -0.002 0.026 0.036 -0.007 0.900 0.754 0.086 0.147 0.104 -0.114 -0.004 0.046 -0.093 0.035 0.006 1.275 0.578 0.182 0.278 0.097 -0.024 0.009 0.036 0.011 0.047 0.002 1.151 0.758 0.219 0.139 0.094 0.037 0.029 0.033 0.031 0.057 0.036 1.177 0.652 0.154 0.248 0.042 0.017 -0.006 0.016 0.002 0.023 -0.030 1.301 0.710 0.146 0.182 0.049 -0.057 -0.035 0.033 0.079 0.065 -0.088 1.365 0.949 0.335 0.466 0.174 -0.033 0.056 -0.064 -0.007 0.051 -0.034 1.485 0.724 0.295 0.167 0.183 0.016 0.027 0.073 0.036 0.057 -0.062 1.099 1.193 0.245 0.204 0.076 -0.014 0.021 0.028 -0.044 0.030 -0.007 0.157 0.336 0.240 0.338 -0.001 0.012 -0.037 0.027 -0.206 -0.002 -0.040 0.083 0.353 0.248 0.151 -0.051 -0.014 -0.036 0.010 0.014 0.040 0.002 0.053 0.086 0.215 0.294 0.021 -0.047 -0.070 -0.021 -0.039 0.089 -0.078 1.110 0.635 0.176 0.144 -0.024 -0.012 -0.039 0.024 -0.056 0.088 0.056 0.243 0.410 0.287 0.056 -0.070 0.020 -0.018 0.001 0.007 0.101 -0.045 0.050 0.385 0.220 0.253 0.029 -0.021 -0.047 -0.010 -0.069 -0.031 -0.067 0.079 0.554 0.038 0.163 0.043 0.163 -0.144 0.046 0.463 0.014 -0.111 0.962 0.666 0.411 0.603 0.371 0.023 0.027 0.063 -0.051 0.042 0.205 0.196 0.677 0.415 0.397 0.001 -0.001 0.029 -0.045 -0.044 0.006 -0.033 0.013 0.219 0.137 -0.084 -0.030 0.031 0.105 0.005 -0.005 0.022 0.000 0.029 0.546 0.086 0.060 55 Table 3.3 Library screen data normalized to local maximum, color coded. Position Building blocks D/E N/Q LVnI R/K S/T X1 F/Y/W H P pS/pT pY Naph-Ala D/E N/Q LVnI R/K S/T X2 F/Y/W H P pS/pT pY Naph-Ala D/E N/Q LVnI R/K S/T X3 F/Y/W H P pS/pT pY Naph-Ala MEG1 PTPe LMWPTP PTPµ CD45 20% 36% 0% 70% 11% 81% 65% 48% 38% 0% 11% 9% 44% 44% 37% 5% 17% 0% 49% 0% 25% 81% 50% 43% 42% 14% 1% 98% 0% 31% 31% 68% 65% 62% 8% 8% 0% 94% 0% 58% 0% 0% 62% 66% 41% 0% 11% 0% 51% 0% 23% 31% 3% 55% 65% 56% 5% 17% 1% 3% 18% 37% 100% 0% 100% 95% 61% 3% 14% 6% 30% 59% 100% 41% 100% 74% 100% 100% 77% 100% 100% 100% 100% 96% 52% 57% 53% 63% 75% 100% 59% 10% 66% 54% 42% 0% 0% 54% 52% 47% 0% 5% 0% 53% 0% 32% 64% 26% 46% 54% 61% 10% 92% 0% 9% 14% 57% 37% 13% 58% 49% 42% 7% 76% 0% 13% 32% 68% 44% 0% 72% 67% 61% 9% 92% 17% 18% 8% 44% 15% 2% 61% 35% 26% 7% 16% 46% 24% 39% 54% 37% 0% 57% 49% 51% 5% 72% 0% 20% 11% 63% 44% 0% 67% 52% 40% 6% 77% 90% 17% 100% 55% 35% 24% 44% 47% 38% 6% 76% 23% 9% 56% 37% 37% 100% 100% 71% 66% 100% 40% 100% 0% 0% 72% 83% 71% 72% 100% 100% 5% 100% 0% 100% 0% 100% 100% 0% 68% 70% 87% 0% 67% 0% 23% 83% 74% 60% 0% 92% 70% 47% 9% 79% 12% 11% 0% 57% 59% 22% 75% 76% 100% 67% 100% 0% 31% 47% 57% 26% 6% 70% 50% 34% 44% 51% 100% 24% 19% 24% 3% 13% 42% 42% 30% 14% 47% 36% 11% 0% 10% 0% 0% 42% 38% 18% 0% 47% 0% 37% 0% 10% 0% 0% 52% 49% 45% 11% 42% 30% 36% 0% 22% 1% 0% 65% 92% 44% 0% 83% 0% 33% 0% 17% 34% 26% 59% 65% 34% 9% 52% 91% 26% 12% 57% 22% 100% 64% 65% 36% 0% 20% 0% 27% 0% 100% 38% 37% 100% 73% 62% 0% 0% 0% 100% 0% 93% 100% 26% 64% 100% 59% 100% 41% 97% 36% 100% 35% 19% 9% 57% 100% 40% 0% 65% 0% 34% 0% 18% Shp2 HePTP VHR TCPTP YopH LYP 71% 11% 44% 79% 79% 25% 70% 60% 69% 16% 19% 38% 11% 0% 0% 56 Table 3.3 Continued Position Building blocks D/E N/Q LVnI R/K S/T X4 F/Y/W H P pS/pT pY Naph-Ala D/E N/Q LVnI R/K S/T X5 F/Y/W H P pS/pT pY Naph-Ala D/E N/Q LVnI R/K S/T X6 F/Y/W H P pS/pT pY Naph-Ala Shp2 HePTP VHR TCPTP YopH LYP MEG1 PTPe LMWPTP PTPµ CD45 32% 29% 55% 48% 73% 13% 100% 8% 17% 7% 57% 25% 20% 49% 46% 62% 8% 55% 7% 37% 0% 66% 29% 9% 62% 53% 50% 11% 66% 6% 5% 4% 57% 27% 29% 36% 41% 33% 1% 26% 100% 25% 0% 55% 38% 13% 70% 88% 81% 4% 57% 7% 88% 0% 98% 25% 12% 63% 55% 40% 9% 55% 51% 3% 2% 73% 34% 20% 38% 64% 47% 7% 86% 0% 29% 4% 48% 21% 28% 38% 31% 56% 0% 0% 22% 33% 0% 87% 95% 100% 64% 93% 100% 100% 58% 0% 100% 0% 100% 100% 23% 100% 100% 81% 43% 63% 0% 23% 100% 88% 18% 0% 86% 60% 61% 0% 69% 0% 8% 0% 30% 82% 92% 49% 89% 100% 100% 100% 0% 0% 0% 86% 34% 0% 52% 86% 90% 0% 0% 7% 69% 0% 76% 77% 0% 53% 81% 53% 0% 55% 0% 7% 38% 61% 19% 0% 54% 51% 25% 0% 52% 0% 0% 46% 85% 18% 0% 51% 78% 65% 0% 14% 21% 100% 0% 80% 50% 0% 100% 35% 46% 0% 37% 0% 0% 0% 69% 64% 40% 47% 100% 72% 0% 32% 0% 36% 100% 83% 0% 0% 62% 22% 29% 0% 35% 0% 0% 0% 81% 58% 100% 57% 81% 96% 0% 83% 0% 0% 0% 69% 100% 98% 58% 77% 62% 89% 31% 0% 0% 0% 100% 70% 0% 88% 54% 33% 0% 48% 100% 0% 0% 35% 61% 0% 35% 82% 88% 100% 100% 0% 56% 9% 25% 4% 30% 50% 50% 45% 58% 23% 0% 58% 0% 13% 0% 0% 14% 33% 17% 15% 23% 8% 15% 0% 9% 0% 0% 41% 28% 28% 0% 0% 100% 100% 100% 5% 0% 0% 25% 40% 38% 0% 29% 15% 47% 0% 13% 0% 0% 28% 27% 39% 0% 53% 0% 48% 0% 3% 21% 51% 46% 38% 29% 25% 28% 0% 20% 45% 23% 30% 0% 41% 61% 36% 0% 0% 0% 21% 0% 100% 100% 15% 100% 100% 100% 16% 52% 0% 48% 0% 27% 55% 100% 36% 28% 33% 3% 36% 0% 30% 0% 7% 0% 3% 30% 15% 11% 0% 0% 4% 20% 0% 3% 57 Table 3.3 Continued Position Building blocks D/E N/Q LVnI R/K S/T X7 F/Y/W H P pS/pT pY Naph-Ala D/E N/Q LVnI R/K S/T X8 F/Y/W H P pS/pT pY Naph-Ala Shp2 HePTP VHR TCPTP YopH LYP MEG1 PTPe LMWPTP PTPµ CD45 25% 0% 45% 43% 66% 0% 13% 0% 36% 48% 68% 82% 0% 75% 65% 51% 14% 46% 0% 93% 4% 100% 0% 0% 52% 60% 81% 0% 43% 0% 0% 0% 7% 5% 0% 63% 26% 32% 0% 0% 32% 56% 0% 50% 56% 0% 48% 54% 60% 0% 63% 0% 55% 17% 70% 47% 0% 64% 65% 30% 16% 50% 14% 73% 6% 64% 43% 100% 55% 46% 53% 53% 45% 39% 88% 100% 65% 28% 53% 60% 44% 39% 0% 22% 2% 35% 0% 72% 48% 0% 80% 100% 100% 0% 45% 100% 100% 0% 75% 100% 0% 61% 88% 36% 100% 0% 0% 79% 0% 82% 94% 36% 100% 73% 44% 48% 100% 46% 87% 0% 61% 22% 0% 50% 58% 56% 20% 44% 0% 30% 0% 14% 6% 5% 52% 60% 25% 0% 43% 0% 0% 0% 7% 0% 0% 13% 52% 49% 0% 16% 3% 40% 1% 5% 12% 0% 94% 42% 24% 0% 0% 0% 88% 0% 100% 0% 0% 61% 69% 9% 0% 38% 0% 87% 27% 22% 0% 9% 57% 53% 42% 0% 2% 2% 100% 0% 5% 9% 0% 82% 9% 27% 0% 0% 0% 0% 0% 7% 19% 100% 98% 99% 100% 0% 74% 100% 14% 0% 87% 100% 15% 100% 100% 66% 26% 100% 0% 41% 100% 18% 3% 0% 32% 33% 0% 28% 0% 0% 6% 0% 1% 0% 26% 81% 21% 10% 100% 9% 0% 22% 0% 3% 58 Table 3.4 A proposed pool of lead sequences. “Bad probe” is a generally bad sequence for most PTPs. Other probes are determined according to their positional preference as concluded from Table 3.3. An alanine is placed when there is not a unique preference at a certain site. Probe Locations X1 X2 X3 X4 X5 X6 X7 X8 Bad probe A A R/K A pCAP P NaphAla L/V/n/I H SHP2 probe P D/E pY H pCAP LVnI pS/pT N/Q A LMWPTP probe R/K A pS/pT S/T pCAP N/Q R/K H F/Y/W PTPmu probe A F/Y/W A pY pCAP H R/K D/E A PTP-Meg1 probe N/Q D/E F/Y/W H pCAP R/K F/Y/W S/T N/Q LYP probe pY P pY A pCAP D/E D/E A NaphAla PTPe probe A P N/Q R/K pCAP NaphAla R/K pS/pT P VHR probe P Naph pS/pT A pCAP F/Y/W pS/pT A H 59 Table 3.5 The sequence of PTPµ probe and LMWPTP probe. Top row: lead sequence candidate summarized from the screen library. Middle rows: a list of natural substrates for LMWPTP (a) and PTPµ (b). Bottom row: the chosen final sequences. a. Probe/ Natural Substrate PTPµ lead sequence candidate LCK, Y394 MK03, Y204 INSR, Y1189, Y1190 TEC, Y519 PGFRB, Y751 PTPµ probe Locations X1 A E F E L E A X3 X4 F/Y/W A D N L pT T D D D S V F A X2 pY E E pY Q D pY X5 X6 X7 X8 H T V R T V H R/K A A K S P K D/E R T G S M E A E R G S L A X5 X6 X7 X8 pCAP N/Q R/K H F/Y/W pCAP pY pY pY pY pY pCAP b. Probe/ Natural Substrate Locations X1 X2 X3 X4 LMWPTP lead sequence candidate R/K A PDGF-r, Y716 S A E L pY S N A L PDGF-r, Y857 Ephrin type-B receptor 1, Y929 LMWPTP probe R D S N pY I S K G K K M A V pT Q S pY pCAP R N D R S H F F pS/pT S/T 60 Figure 3.6 Enzymatic reactivities of the probes. a: A comparison of substrate reactivities between the general bad probe and the PTPµ probe. b: A comparison on substrate reactivities between the general bad probe and the LMWPTP probe. Table 3.6 The final yields of the purified CPP library. The masses of the peptides were verified with LC-MS. # Peptide sequence MW yield (mg) % Yield 1 Ac-HRLRRS-bAla-KFluor-CONH2 1422.29 30.81 72% 2 Ac-SAYHRLRRS-bAla-KFluor-CONH2 1743.62 17.06 33% 3 Ac-HRLRRC-bAla-KFluor-CONH2 1438.35 31.93 74% 4 Ac-CAYHRLRRC-bAla-KFluor-CONH2 1775.74 28.14 53% 5 Ac-GRKKRRQRRRPPQ-bAla-KFluorCONH2 2317.37 23.05 33% 6 Ac-R4 -bAla-KFluor-CONH2 1241.10 30.96 83% 7 Ac-R5-bAla-KFluor-CONH2 1397.29 23.53 56% 8 Ac-R6-bAla-KFluor-CONH2 1553.47 32.13 69% 9 Ac-R7-bAla-KFluor-CONH2 1709.66 31.23 61% 10 C12-R7-bAla-KFluor-CONH2 1849.82 25.39 46% 11 C14-R7-bAla-KFluor-CONH2 1877.85 41.30 73% 12 C16-R7-bAla-KFluor-CONH2 1905.88 22.43 39% 13 Ac-R8-bAla-KFluor-CONH2 1865.85 24.09 43% CHAPTER 4 ACTIVITY-BASED PEPTIDE PROBES FOR TARGETING SPECIFIC PTPS 4.1 Introduction PTPN22, or lymphoid-specific tyrosine phosphatase, is primarily expressed in lymphoid tissue and acts as a downregulator of the T-cell activation pathway.1 It is one of the most studied enzymes among the PTP superfamily due to its intriguing role in the immune system.2 Studies have demonstrated that a human variant R620W singlenucleotide polymorphism in PTPN22 gene is associated with autoimmune disorders, leading to a significantly increased risk of type-1 diabetes, rheumatoid arthritis, and systemic lupus erythematosus.3 In addition, a gene knock-out study has indicated that PTPN22 may act as a positive regulator of anaphylaxis4. Obiri et al observed that PTPN22 knock-out mice generally showed a reduced level of inflammation compared to the wild-type mice, which indicates that PTPN22 helps facilitate the allergic reaction. When a PEP inhibitor C28 is co-administrated with other anti-inflammation therapy, the repression of mast cells turned out to be more efficient.4 These observations all suggested that PTPN22 inhibition may evolve as a new mechanism for treating allergic reactions, inflammation, and autoimmune diseases.2,5 Despite the potential therapeutic importance of PTPN22, its detailed biological function and mechanism in multiple immune signaling pathway remain in mystery. In 62 order to develop a more thorough understanding of PTPN22, useful research tools are always in great need. The selection of current research probes is quite limited, as most of the existing probes are nonselective, such as DiFMUP and pNPP. These pan-specific substrates can be hydrolyzed by most PTPs as well as other protein phosphatases. Only a few PTPN22 inhibitors with sub-micromolar potencies have been reported, and often the selectivity is ambiguous.6–8 Chemical probes with an affinity handle or a fluorescent tag would be preferential since they can be used as a pull-down handle to provide additional information. To address these limitations of the current chemical probes and provide useful tools to facilitate biological studies of PTPN22, we designed a PTPN22-selective activity-based probe as illustrated in Figure 4.1. The design of our activity-based probe consists of a suicide inhibitor which will react with the catalytic cysteine, a peptide chain which can interact with PTPN22 to provide specificity, and an affinity tag for labeling or pull-down of the enzyme-probe adduct. 4.2 Results and discussion To identify a selective peptide for targeting PTPN22, a randomized library of peptide sequences was designed and screened by Dr. Ryan Mathews in our lab using a One-Bead-One-Compound technique (unpublished work). The two most promising PTPN22-selective sequences identified from the library, LDLL-X-SDDD and EDNE-XDARE (X represents the position of pCAP), were incorporated into the probe. Aiming for an easy, modular-based incorporation of the warhead molecule into the probe, “click” chemistry was used for the probe assembly. The X position was replaced with a propargyl-glycine residue, and an azide functional group was incorporated in the warhead 63 motif. Liu et al previously reported a series of phenylvinyl sulfonate compounds to be pan-specific covalent PTP inhibitors at a low milimolar range,9 which appears to be an ideal warhead molecule for our probe. The synthesis of 4-(azidomethyl) phenylvinyl sulfonate was achieved in three steps starting from the commercially available 4hydroxyl benzyl phenol (Scheme 4.1). After selective bromination on the benzyl alcohol, the bromine group was substituted by an azide group, followed by a coupling reaction between the azide and the 2-chloroethanesulfonyl chloride to give the final product, which was purified and characterized using 1H, 13C NMR, and LC-MS. The warhead moiety, BzVsO, was then installed in the two peptide sequences (Ac-EDNEpropargylglycine-DARE-NH2 and Ac-LDLL- propargylglycine-SDDD-NH2) via “click” chemistry (Scheme 4.2). The selective inhibition of PTPN22 by the two probes was tested using a panel of PTPs. As expected, the two probes inhibited PTPN22 activity in a concentration- and time-dependent manner (Figure 4.2a). The IC50 values of the EDNE and LDLL probes after 1 h of incubation with PTPN22 (0.39 ± 0.07 mM and 0.16 ± 0.01 mM, respectively), were similar to that of the warhead molecule BzVsO alone (IC50 = 0.17 ± 0.02 mM under the same conditions). To determine the PTPN22-selectivity of the two probes, the IC50 values towards a series of PTPs were also measured and summarized in Figure 4.3 and Table 4.1. Both probes exhibit moderate selectivity over PTP1B, PTP-PEST, TCPTP, and YopH, but they also inhibit SHP2 and HePTP with similar potency. The EDNE probe (IC50 = 0.130 ± 0.006 mM) is less potent than the LDLL probe (IC50 = 0.08 ± 0.01 mM) but more selective towards PTPN22. Covalent inhibitors such as BzVsO often follow a two-step enzyme inactivation 64 mechanism with an initial, reversible, binding event followed by a slow covalent inactivation step.10 To characterize the covalent inhibition parameters for the two covalent inhibitors, a more detailed kinetic analysis was carried out as described previously.10 The pseudo-first-order inactivation rate constant (kobs) for the covalent interaction was fitted from the time-dependent inhibition curve (Figure 4.4 a & b). The kobs was then plotted with probe concentration to determine the inactivation constant (kinact) and apparent dissociation constant for the reversible complex (Kapp). The LDLL probe follows the classic covalent inhibition pattern and fitting of Kobs gives the kinact value as 0.00107± 0.00005 S-1. Interestingly, the EDNE probe does not follow the classic pattern. Instead, at high probe concentrations, the inhibition efficiency seems to decrease, indicating potential self-inhibitory effects. KI, the inhibitor dissociation constant, was calculated from the Kapp and Km,DiFMUP (20 µM, experimentally determined) as previously described.10 As a comparison, the kinact and KI values of two other reported covalent inhibitors are listed in Figure 4.4 e. The inactivation constant kinact and inhibitor dissociation constant of the probe remain similar to the warhead compound PVSN, which are significantly superior compared to the commercially available competitive inhibitor a-bromobenzyl phosphonate. Once the inhibition motif and selective peptide sequence were determined, a biotin handle was installed to serve as an affinity tag to facilitate the identification of covalently labeled enzyme. A lysine residue with allyloxycarbonyl (Alloc) protected side chain was coupled to the N-terminus of the peptide, followed by Pd(PPh3)4 catalyzed deprotection and biotinylation. The purified biotinylated peptides were then “click” assembled with freshly synthesized 4-azido-phenyl-vinyl sulfonate as described in Scheme 4.3. The potencies of the biotinylated probes were verified again with PTPN22 65 (Figure 4.5). The potency of the EDNE probe remains unchanged, while the potency of the LDLL probe dropped at least 2-fold after addition of the biotin tag. The biotinylated EDNE probe was chosen to proceed with future experiments based on its better selectivity and higher potency compared to the biotinylated LDLL probe. To demonstrate that the biotinylated EDNE probe is truly “activity-based”, the probe binding efficiencies with enzymes at different activity levels were tested. It is known that in the presence of excess hydrogen peroxide in vitro, the catalytic cysteine C227 in PTPN22 forms a disulfide bond with a back-door cysteine C129 and becomes inactivated.11 Therefore, increasing concentrations of H2O2 were used to oxidatively inactivate PTPN22. The biotinylated EDNE probe was incubated with the oxidized PTPN22 for 1 h at rt to form the covalent enzyme probe-adduct. An equal amount of enzyme from different treatment groups was loaded in each well of an SDS-PAGE gel and subject to immunoblotting using HRP cross-linked anti-biotin. The presence of enzyme was revealed by the total protein stain Ponceau S, while the amount of probe bound to active enzyme was revealed using HRP cross-linked anti-biotin. As shown in Figure 4.6, abrogation of labeling was clearly observed with the H2O2 inactivated PTPN22. It is worth mentioning that some probe-enzyme adduct was still observed when the enzyme has been fully inactivated with 10 mM of H2O2, indicating potential nonspecific binding between the inactivated enzyme and the probe. We further explored the possibility of detecting active PTPN22 in a more complex biological environment using the biotinylated EDNE probe. PTPN22-expressing E.coli cells were prepared, lysed, and incubated with the biotinylated EDNE probe. As shown in Figure 4.7, PTPN22 was successfully labeled by the probe with minimal offtarget labeling, despite the presence of numerous cysteine-containing proteins and 66 enzymes in the lysate. This demonstrated the ability of the ABP to serve as a pull-down handle in cellular experiments. It is worth noting that an intense band was observed around 20 kDa in all the PTPN22 expressing E.coli lysates tested. To understand where the false positive signal came from, we examined two control lanes using the PTPN22 expressing E. coli lysate without incubation of the biotinylated EDNE probe. We still observed the 20 kDa band in the absence of the probe, which indicated that this intense signal resulted from an unknown protein binding to the anti-biotin antibody, not off-target labeling by our probe. An interesting observation in the previous selectivity experiment was the EDNE probe’s high potency towards SHP2. Since the EDNE probe has a 10-fold higher potency towards SHP2 than LYP (Table 4.1), we explored the possibility of applying the EDNE probe as a SHP2-specific probe. Indeed, the EDNE probe exhibits excellent specificity towards SHP2 in the SHP2-expressing E.coli lysate with little off target labeling (Figure 4.8 c). H2O2 treated, purified SHP2 was then tested with the biotinylated EDNE probe to determine whether the labeling efficiency is dependent on the SHP2 activity (Figure 4.8). To our surprise, the abrogation of labeling did not occur until the SHP2 was treated with 20 mM of H2O2, much more than the 10 mM that is required to fully inactivate the enzyme (Figure 4.8 b). Since H2O2 is a nonspecific oxidant, it can potentially oxidize critical structural residues (Cys, Met) outside the catalytic pocket and cause SHP2 to lose enzymatic activity. Under such conditions, the catalytic cysteine of the inactivated enzyme can still react with the biotinylated EDNE probe and give a false positive result. To test this hypothesis, we changed the oxidant to pervanadate, an inhibitor known to specifically oxidize the catalytic cysteine with an IC50 of 14 µM.12 To our surprise, the amount of 67 enzyme-EDNE probe adduct did not decrease as SHP2 loss enzymatic activity. Rather, a higher level of enzyme-probe adduct was observed (Figure 4.9). This result indicated that the EDNE probe might have conjugated with a nonactive site cysteine and inhibited SHP2 in an allosteric manner. To further investigate which site was conjugated to the EDNE probe, we conducted a probe mapping experiment. SHP2 was incubated with an excess amount of biotinylated EDNE probe until fully inhibited as determined by a coupled DiFMUP assay. After removing excess EDNE probe through a desalting step, the probe-enzyme adduct was digested with trypsin and subject to LC-MS/MS analysis to map the actual binding site. Indeed, the catalytic site cysteine (C459) was found mostly unconjugated, while C333 and C318 were both found to be predominantly conjugated by the EDNE probe (Table 4.2). C318 is located in a flexible loop, and has not been reported to be structurally important to the enzymatic activity. The C333 site is also away from the catalytic site. However, Chio et al have recently reported C333 to be an allosteric inhibition site for biarsenical compounds.13 This is an interesting finding as the C333 position is usually occupied by proline in other classic PTPs.13 Thus, the EDNE probe will serve as a SHP2-specific, allosteric-site targeting probe to selectively knockdown and label SHP2 in biological samples. 4.3 Conclusion An activity-based probe was designed and synthesized. The covalent inhibition profile of PTPN22 was examined, and the selectivity was validated with a list of PTPs. However, the probe was found to be a more potent inhibitor of SHP2 and exhibited decent specificity towards SHP2 in the presence of other cysteine-containing proteins in 68 an E.coli lysate, potentially conjugated to an allosteric site. The inhibition of SHP2 is highly selective, but not activity dependent. The mechanism of this allosteric inhibition is not currently understood. 4.4 Experimental section All reagents and organic solvents were obtained from commercial sources and used without further purification. All NMR spectra were recorded on a Mercury 400 MHz spectrometer and were referenced to residual solvent peaks or TMS (d 0.00 ppm). Peptide masses were determined on a micro MX Maldi-ToF mass spectrometer. Column chromatography was performed on EMD silica gel (60-200 mesh) or EMD alumina (80200 mesh). PTPN22 and SHP2 enzymes were expressed and purified as described in Chapter 2. 4.4.1 Synthesis of 4-(bromomethyl) phenol Cyanuric chloride (0.92 g, 5 mmol) was suspended in 1 mL of DMF and stirred for 3 h at rt until the formation of an off white solid, which was then suspended in 12.5 mL of DCM and added with NaBr (0.98 g, 9.5 mmol). The mixture was stirred for 10 h, followed by the addition of 4-(hydroxymethyl) phenol (0.94 g, 4.7 mmol). The resulting mixture was further stirred for 5 h and monitored by TLC until the reaction was completed. The reaction mixture was then filtered through celite. The organic phase was washed with H2O, 1 N HCl solution, brine, dried over Na2SO4, and concentrated in vacuo to give 4-(bromomethyl) phenol as a yellow solid (yield: 49%). 1H NMR (400 MHz, CDCl3) d (ppm) 8.30 (s, 1H), 7.43 (m, 2H), 7.14 (m, 2H), 4.58 (s, 2H). 13C NMR (100 MHz, CDCl3) d (ppm) 159.22, 135.93, 130.22, 121.73, 45.57. It is worth noting that the 69 NMR spectrum indicated a mixture of 4-(bromomethyl) phenol and 4-(chloromethyl) phenol. However, they were both converted into the 4-(azidomethyl) phenol in the next step. 4.4.2 Synthesis of 4-(azidomethyl) phenol To a stirred solution of 4-(bromomethyl) phenol (310.70 mg, 1.66 mmol) in DMF was added sodium azide (129.50mg, 1.99 mmol) and the resulting solution was stirred for 6 h. After the reaction was completed, the mixture was diluted with the ethyl acetate and washed with brine, dried over Na2SO4, filtered through a pad of silica, and concentrated to give 4-(azidomethyl) phenol as a clear oil (yield: 69%). 1H NMR (400 MHz, CDCl3) d (ppm) 7.22 (d, 2H), 6.91 (d, 2H), 4.29 (s, 2H). 13C NMR (100 MHz, CDCl3) d (ppm) 156.72, 132.18, 130.04, 115.87, 54.60. 4.4.3 Synthesis of 4-(azidomethyl) phenyl vinylsulfonate To a stirred solution of 4-(azidomethyl) phenol (170.9 mg, 1.146 mmol) in 10 mL of anhydrous DCM was added triethylamine (417.55 mg, 4.126 mmol) and cooled down to 0 °C under N2. 2-Chloroethanesulfonyl chloride (224.16 mg, 1.375 mmol) was then added dropwise and the reaction mixture was allowed to gradually warm to rt and stir overnight. After the reaction was completed, the mixture was diluted with DCM, washed with chilled brine, dried over Na2SO4, and purified through alumina column chromatography to afford 4-(azidomethyl) phenyl vinylsulfonate as a brown oil (yield: 92%). 1H NMR (400 MHz, CDCl3) d (ppm) 7.34 (m, 2H), 7.25 (m, 2H), 6.67 (dd, 1H), 6.37 (d, 1H), 6.18(d, 1H). 13C NMR (100 MHz, CDCl3) d (ppm) 149.36, 135.06, 132.20, 130.19, 129.79, 122.93, 54.15. 70 4.4.4 General solid phase peptide synthesis Peptides were synthesized on an AAPTech Focus XC automated peptide synthesizer using rink amide resin (0.69 mmol/g). The Fmoc group on resin and amino acids were cleaved with 20% piperidine for 5 min three times, and the typical coupling reaction was carried out using 3 equiv. of Fmoc-amino acids, 3 equiv. of DICI, 5 equivalence of HOBT, in anhydrous DMF and was allowed to react at rt with shaking for 90 min. 4.4.5 Solid phase peptide synthesis of biotinylated peptides 100 µmol peptide was synthesized on Rink Amide resin (0.4 mmol/g) using the same protocol on synthesizer until the completed coupling of Fmoc-Lys(alloc)-OH. The peptides were transferred to a manual synthesis vessel and re-suspended in 2 mL of anhydrous DCM. Twenty-four equiv. of PhSiH3 were added to the suspension and allowed to stir for 5 min, followed by the addition of 0.25 equiv. Pd(PPh3)4 under argon. The reaction was allowed to proceed for 30 min, then washed with DCM and repeated with fresh reagents again for 30 min. After the second reaction was completed, the resins were washed extensively with DCM, 0.2% TFA/DCM, 5% DIPEA/DCM, and finally with DMF. The resin was re-suspended in anhydrous DMF, then added with 2 equiv. dbiotin, and 2 equiv. PyBOP, 4 equiv. DIPEA and shaken vigorously for 1 h. The completion of each coupling reaction was monitored by ninhydrin test. Peptide cleavage and ether precipitation: the synthesized peptides were cleaved from the resin and deprotected using a cleavage cocktail containing 88% TFA (v/v), 5% phenol (v/v), 5% water (v/v), and 2% Triisopropylsilane (v/v) for 6 h at rt. The cleavage solution was slowly precipitated in 50 mL of cold ether and incubated in freezer 71 overnight. The precipitated peptides were collected through centrifugation and washed with 2 x 50 mL of cold ether, air-dried, and purified on reverse phase HPLC when necessary. The pure peptide or probe was characterized by MOLDI-TOF. The observed peptide masses are listed below. • Ac-LDLL-(Pra)-SDDD-NH2: MS calculated: 1040.5, MS observed: 1063.5 [M+Na+] • Ac-LDLL-(BzVsO)-SDDD-NH2: MS calculated: 1279.7, MS observed: 1280.3 [M+H+] • Ac-EDNE-(BzVsO)-DARE-NH2: MS calculated: 1351.7, MS observed: 1352.3 [M+H+] • Ac-Kbiotin-(bAla)-EDNE-BzVsO-DARE-NH2, MS calculated: 1776.9, MS observed:1777.2 [M+H+] • Ac-Kbiotin-(bAla)-LDLL-BzVsO-SDDD-NH2, MS calculated: 1704.9, MS observed: 1706.7 [M+H+] 4.4.6 In vitro enzyme inhibition assay Enzyme activity was tested using standard DiFMUP assay at rt on a Molecular Devices Spectramax M5 multimode plate reader with excitation and emission at 350 and 455 nm, respectively. The final concentration of PTP enzymes used for in vitro assays was 1 nM. The final concentration of the DiFMUP used was 1.5 μM. The total DMSO concentration in each reaction was held constant at 6% of the total reaction volume (100 μL). Stock solutions and serial dilutions of the compounds for determination of IC50 were made in DMSO. PTP enzymes are pre-activated by 1 mM TCEP for 30 min prior to activity assays. 72 4.4.7 Measuring IC50 The probes at various concentrations were incubated with the enzyme for 60 min before initiating the activity assay by adding DiFMUP as the fluorescent substrate. The percent relative activity for each compound was determined by factoring the measured fluorescence values for the compound treated wells over the control wells treated with DMSO only. The resulting plot of inhibitor concentration versus percent enzyme activity provided the IC50 values. 4.4.8 KI and kobs determination The probes at various concentrations were added to the wells containing the assay buffer. DiFMUP was added to the wells, followed by the pre-activated enzyme immediately to initiate the reaction. Enzyme activity was tested on a Molecular Devices Spectramax Gemini XS plate reader with excitation and emission at 350 and 455 nm, respectively. The observed first-order rate constant (kobs) was calculated for each inhibitor concentration using the method reported in reference.10 The KI, kinact are calculated by fitting kobs over probe concentration using the equation described in reference.10 4.4.9 E.coli BL21 lysate preparation An aliquot of E.coli BL21(DE3) competent cells was transformed with 1 μL of plasmid solution. The successful transformants were selected by respective antibiotic agar plates, and subsequently inoculated into 10 mL overnight starter LB culture. The starter culture was inoculated into 250 mL of LB medium and cultured in 30 ºC shaker until OD600 reaches 1.0, then 1mM IPTG was added to induce protein expression at 18 ºC. The cell culture flasks were kept shaking at 18 ºC overnight and harvest by 73 centrifugation. The cell pellet was re-suspended in 10 mL of lysis buffer, added with PMSF (final conc. 1 mM), BME (final conc. 10 mM), and lysozyme (final conc. 0.6 mg/mL), incubated on ice for 30 min and followed by sonication for 2 x 1.5 min cycles. Triton 1% solution (10µL/mL) was added in between the two cycles and incubated for 10 min on ice. The cell lysate was then centrifuged at 14000 rpm for 50 min. After centrifuge, the supernatant was clarified through 0.22 µm filter, aliquoted, and flashfrozen immediately. 4.4.10 Western blot Cell lysate samples were resolved by electrophoresis using a 10% SDS-PAGE mini gel. After electrophoresis, the gel was blotted with a nitrocellulose membrane and placed in a Bio-rad Semi-Dry Transfer Cell to transfer at 25 V, 400mA for 1 h. To determine the transfer efficiency, the membrane is stained with Ponceaun S solution, and the gel is stained with Coomassie Brilliant Blue G-250 Dye. The membrane was rinsed 2 to 3 times with water and TBST buffer to eliminate Ponceau S stain, and subsequently blocked with fresh 5% BSA in TBST for 1 h at rt, or 3 h at 4 ºC. The anti-biotin antibody was incubated with the membrane at 1:10000 dilutions overnight at 4 ºC (other antibodies were used following manufacturer’s instruction). The membrane was washed with TBST buffer (10 min ´ 4), then incubated with western ECL-blotting substrate (Bio-Rad #1705062) for 5 min and imaged with Bio-Rad imaging system. The antibody can be stripped off by incubating the membrane in standard stripping buffer (0.2 M glycine, 0.1% SDS, 1% Tween 20, pH 2.2) for 10 min twice and subsequently rinsed twice by 10 min PBS, 5 min TBST buffer. 74 4.4.11 Trypsin digestion and probe conjugation mapping experiment An aliquot of SHP2 was concentrated to 1 mg/mL using a Nanosep centrifugal device (10K MWCO, Pall, #Z722065). A 75 µL aliquot of the concentrated enzyme was incubated with excess of the biotinylated EDNE probe and monitored by assaying aliquot of enzymes using DiFMUP. The probe-treated enzyme was incubated with 10 mM DTT in 50 mM ammonium bicarbonate (freshly prepared) for 40 min at rt on a shaker to reduce the disulfide bonds. The reduced enzyme was alkylated with 20 mM iodoacetamide in 50 mM ammonium bicarbonate for 30 min in the dark at rt on a shaker. The sample was reduced again with 10 mM DTT to eliminate excess iodoacetamide. After that, the sample was placed in a Microsep centrifugal device (10K MWCO, Pall, # MCP100C41) to desalt and eliminate the excess amount of biotinylated EDNE probe. 10% Acetonitrile was added to the desalted sample to help solubilize the protein. Lyophilized Trypsin powder (Sigma Life Science, # SLBG6452V) was dissolved in 50 mM acetic acid (1 mg/mL) and added to the sample in a ratio of enzyme: trypsin = 20:1. The reaction was placed in a 37 ºC shaker overnight. The sample was then analyzed using LC-MS/MS measured on a Xevo G2-XS Q-Tof with Acquity I Class UPLC. 4.5 Reference (1) Bottini, N.; Vang, T.; Cucca, F.; Mustelin, T. Semin. Immunol. 2006, 18, 207. (2) He, R.-J.; Yu, Z.-H.; Zhang, R.-Y.; Zhang, Z.-Y. Acta Pharmacol. Sin. 2014, 35, 1227. (3) Fousteri, G.; Liossis, S.-N. C.; Battaglia, M. Clin. Immunol. 2013, 149, 556. (4) Obiri, D. D.; Flink, N.; Maier, J. V; Neeb, A.; Maddalo, D.; Thiele, W.; Menon, A.; Stassen, M.; Kulkarni, R. a; Garabedian, M. J.; Barrios, A. M.; Cato, a C. B. Allergy. 2012, 67, 175. 75 (5) Vang, T.; Congia, M.; Macis, M. D.; Musumeci, L.; Orrú, V.; Zavattari, P.; Nika, K.; Tautz, L.; Taskén, K.; Cucca, F.; Mustelin, T.; Bottini, N. Nat. Genet. 2005, 37, 1317. (6) Yu, X.; Sun, J.-P.; He, Y.; Guo, X.; Liu, S.; Zhou, B.; Hudmon, A.; Zhang, Z.-Y. Proc. Natl. Acad. Sci. U. S. A. 2007, 104, 19767. (7) Hou, X.; Li, R.; Li, K.; Yu, X. J. Med. Chem. 2014, 57, 9309. (8) Thorson, M. K.; Puerta, D. T.; Cohen, S. M.; Barrios, A. M. Bioorg. Med. Chem. Lett. 2014, 24, 4019. (9) Liu, S.; Zhou, B.; Yang, H.; He, Y.; Jiang, Z.-X.; Kumar, S.; Wu, L.; Zhang, Z.-Y. J. Am. Chem. Soc. 2008, 130, 8251. (10) Ahmed, V. F.; Bottini, N.; Barrios, A. M. ChemMedChem. 2014, 9, 296. (11) Tsai, S. J.; Sen, U.; Zhao, L.; Greenleaf, W. B.; Dasgupta, J.; Fiorillo, E.; Orrú, V.; Bottini, N.; Chen, X. S. Biochemistry 2009, 48, 4838. (12) Huyer, G.; Liu, S.; Kelly, J.; Moffat, J.; Payette, P.; Kennedy, B.; Tsaprailis, G.; Gresser, M. J.; Ramachandran, C. J. Biol. Chem. 1997, 272, 843. (13) Chio, C. M.; Lim, C. S.; Bishop, A. C. Biochemistry 2015, 54, 497. 76 Figure 4.1 An illustration of the activity-based probe. Br OH DMF, cyanuric acid N3 O O Cl NaN3, DMF, rt N3 S Cl TEA, DCM 0°C to rt NaBr, rt OH OH OH O O S O Scheme 4.1 Synthesis of 4-(azidomethyl) phenylvinyl sulfonate. N3 Ac-EDNE H N O O DARE-amide DARE-amide HN Ac-EDNE 0.2 eq CuSO4 N O O S O 4 N N 0.4 eq sodium ascorbate EDNE probe O O S O Scheme 4.2 Click assembly of the ABPs illustrated with the EDNE probe sequence. 77 a. b. c. Figure 4.2 Comparing the potency of LDLL probe and the EDNE probe. a. Timedependent inhibition of LDLL probe. b&c. Dose-dependent inhibition of LDLL probe and ENDE probe 78 PTP inhibition assay with Ac-EDNE-BzVsO-DARE probe 110 100 % Activity 90 80 70 60 LYP PTP-PEST PTP1B TCPTP YopH 50 40 0 100 200 300 400 500 600 [Probe] µM Figure 4.3 Inhibition assay tested on a panel of PTP, showing a moderate selectivity towards LYP over some common PTPs (probe was incubated with the enzymes for 15 min). Table 4.1 The selectivity of EDNE probe and LDLL probe. TCEP pre-activated enzymes (1nM) were incubated with the probes for 60 min before addition of the substrate to test the activity. EDNE probe IC50 (mM) LYP TCPTP VHR CD45 PTP1B SHP2 HePTP 0.39 ± 0.07 >1.6 >1.6 >1.6 1.001 ± 0.002 0.050 ± 0.009 0.119 ± 0.02 Selective by fold >4.1 >4.1 >4.1 2.6 0.1 0.3 LDLL probe IC50 (mM) 0.16 ± 0.01 >1.6 0.722 ± 0.005 0.026 ± 0.004 0.382 ± 0.005 0.020 ± 0.002 0.078 ± 0.005 Selective by fold >10 4.5 0.2 2.4 0.1 0.5 79 a. b. c. d. e. Figure 4.4 Determination of covalent inhibition kinetic constants. a&b. Time-dependent covalent inhibition of the LDLL probe and EDNE probe. c&d. Fitting kobs into the inactivation constant kinact and apparent dissociation constant Kapp. e. A comparison of the kinetic constants between the two probes and two existing covalent inhibitors, PVSN is the warhead molecule. 80 i. Fmoc ii. NH2-βAla-EDNE-PrG-DARE biotin Ac-Lys-βAla-EDNE-PrG-DARE vi. biotin Ac-Lys-βAla-EDNE-PrG-DARE + N3 O O S O vii. v. Alloc Fmoc-Lys-βAla-EDNE-PrG-DARE iii, iv biotin Fmoc-Lys-βAla-EDNE-PrG-DARE biotin Ac-Lys-βAla-EDNE-BzVsO-DARE Scheme 4.3 The synthesis of biotinylated probes (illustrated with EDNE probe). i. 20% PIP deprotection, 3 eqv DICI coupling. ii. 3 eqv DICI. iii. 0.25 eqv Pd(PPh3)4, 24 eqv PhSiH3, DCM, Ar. iv. 2 eqv Biotin, 2 eqv PyBOP, 4 eqv DIPEA. v. 3 eqv acetic acid, 5eqv HCTU 10 eqv DIPEA. vi. cleave with TFA: phenol: H2O: TIPS = 88:5:5:2, precipitate from cold ether and dried. vii. 1 eqv copper sulfate (pentahydrate), 2 eqv sodium ascorbate, 2 eqv aminoguanidium-HCl, in THF/H2O/DMSO:5/4/1. Azide:alkyne = 2:1. 81 Figure 4.5 The inhibition test of the activity-based probes on LYP. Ponceau S stain Anti-biotin Figure 4.6 PTPN22 treated with H2O2 and incubated with biotinylated probe. 82 Purified PTPN22 PTPN22 expressing E.coli lysate PTPN22 expressing E. coli (duplicate) - PTPN22 - Non-specific protein binding to anti-biotin Incubated with biotinylated EDNE probe Figure 4.7 ENDE probe detecting PTPN22 in the PTPN22-expressing E.coli lysate. Printed: 4/4/17 1:10 PM Printed: 4/4/17 1:10 PM Page 1 of 1 Page 1 of 1 83 a. c. b. Figure 4.8 EDNE probe detecting SHP2 in biological samples. a. Incubation of probe with H2O2 deactivated SHP2 and corresponding probe labeling (top = Ponceau S stain, middle = anti-biotin-HRP, bottom = antiSHP2). b. Enzyme activity tested using DiFMUP c. Lysate from E. coli overexpressing SHP-2 was incubated with Kbiotin-EDNE-BzVSO-DARENH2 for 60 min at room temperature, resolved on SDS-PAGE gel, and transferred to a nitrocellulose membrane, which was blotted with antibiotin-HRP antibody. The lane on the left shows the total protein content of the lysate by Ponceau S stain while the lane on the right shows only probe-labeled proteins (SHP2). 84 a. Conditions Pervanadate conc. (µM) Incubation time (min) 1 0 90 2 1 90 3 10 90 4 50 90 5 100 90 Ponceaun S stain Kb-EDNE probe/Anti-biotin Anti-SHP2 b. Figure 4.9 Determination of the EDNE probe labeling efficiency related to enzyme activity. a. Incubation of probe with pervanadate deactivated SHP2 and corresponding probe labeling (top = Ponceau S stain, middle = antibiotin-HRP, bottom = anti-SHP2). b. Enzyme activity tested using DiFMUP 85 Table 4.2 Summary of the SHP2 trypsin digestion/probe mapping experiment result. The corresponding cysteine residues were highlighted. Site Trypsin digestion fragment Modifier Modifier calculated mass (Da) Calculated fragment mass (Da) Observed Mass (Da) C318 CNNSKPK Ac-Kbiotin-bA -EDNE-BzVsO-DARE 1777.85 2567.0713 2567.0481 CNNSKPK Ac-Kbiotin-bA -EDNE-BzVsO-DAR 1649.72 2439.0127 2438.9868 Ac-Kbiotin-bA -EDNE-BzVsO-DARE 1777.85 3892.6753 3892.6824 Ac-Kbiotin-bA -EDNE-BzVsO-DARE 1777.85 4020.7703 4020.7463 Ac-Kbiotin-bA -EDNE-BzVsO-DAR 1649.72 3764.6167 3764.5999 Iodoacetamide 57.05 2083.0059 2082.9594 C318 C333 C333 C333 C459 SYIATQGCLQNT VNDFWR KSYIATQGCLQN TVNDFWR SYIATQGCLQNT VNDFWR QESIMDAGPVVV HCSAGIGR CHAPTER 5 DUAL COLORIMETRIC AND FLUOROGENIC PROBES FOR VISUALIZING TYROSINE PHOSPHATASE ACTIVITY S. Biswas, B. S. McCullough, E. S. Ma, D. LaJoie, C. W. Russell, D. Garrett Brown, J. L. Round, K. S. Ullman, M. A. Mulvey and A. M. Barrios, Chem. Commun., 2017, 53, 2233. Reproduced with permission from The Royal Society of Chemistry. 87 88 89 90