Development of a Small Molecule C-X-C Chemokine Receptor Type 4 (CXCR4) Agonist

Authorship: Tiffany Ding*, Susriya Gangireddy*, Andrew Shum, Richard J. Miller** *These authors contributed equally **Corresponding author: r-miller10@northwestern.edu Student: Illinois Math and Science Academy, 1500 Sullivan Road, Aurora, IL Mentor: Northwestern University-Feinberg School of Medicine, 645 N Michigan Avenue, Suite 900, Chicago, IL Abstract: Activation of the C-X-C chemokine receptor type 4 (CXCR4) by its natural ligand stromal cell-derived factor 1 (SDF-1) is known to be involved in a number of vital physiological processes such as inflammation, cell migration, and tumorigenesis. Although there have been a number of antagonists identified, there are no known natural small molecule agonists for the CXCR4 receptor. NUCC-54118 was synthesized to act as an agonist using an in silico receptor binding simulation based on the crystal structure of the CXCR4 receptor. Calcium imaging was used to assess the effect of 54118 on human CXCR4 receptors expressed in a cultured melanoma cell line. Our findings conclude that small molecule 54118 dose-dependently activated a calcium response in these cells. These effects were inhibited by the selective CXCR4 antagonist, AMD3100. 54118 appears to represent the prototype of a novel family of small molecule agonists at the CXCR4 receptor and could have many important uses in medicine, such as the mobilization of hematopoietic stem cells for transplant purposes. Introduction: Many receptors are proteins on cell membranes that receive stimuli from the environment and whose activation induces signal transduction within the cell. These external stimuli include signals from neighboring cells, which allows for the establishment of extensive intercellular communication networks. Receptors are stimulus-specific, meaning they only allow certain molecules or proteins to bind and activate intracellular signaling pathways (Zlotnik, et al. 2006). Chemokines are small, secreted proteins which bind to seven-transmembrane G-protein coupled receptors (GPCR) on cells to induce chemotaxis and regulate the body’s inflammatory responses (Miller, et al. 2008; Kroeze, et al. 2015). The C-X-C chemokine receptor type 4 (CXCR4) is one of 18 known chemokine receptors in the human body (Zlotnik, et al. 2006). The CXCR4 receptor is found on a variety of cell types, including stem and cancer cells, and plays a role in several diseases. For example, it acts as the co-receptor for HIV-1 entry into target cells (Petit, et al. 2007). The endogenous unique ligand for the CXCR4 receptor is the chemokine stromal cell-derived factor 1 (SDF-1) or CXCL12. In response to infection, the release of chemokines from the site of inflammation will guide white blood cells to that site as part of the inflammatory response. This highlights the potential for CXCR4 agonists to assist in important physiological roles, like in stem cell migration and organ transplants (Lewellis and Knaut 2012). SDF-1, however, can have adverse effects. A prevalent example is demonstrated in cancer cells, which rely heavily on the activation of their CXCR4 receptors to induce tumorigenesis and cancer cell metastasis (Banisadr, et al. 2011; Murdoch 2000). AMD3100 is a CXCR4 antagonist, a molecule that blocks activation of the CXCR4 receptor. In theory, antagonists for the CXCR4 receptor could have important therapeutic uses. For example, blocking CXCR4 receptors expressed on cancer cells could help in preventing cancer cell metastasis. There have been many robust CXCR4 antagonists discovered, which serve a wide range of therapeutic purposes, including the mobilization of hematopoietic stem cells for transplantation purposes and inhibition of HIV-1 infectivity (Liang, et al. 2012; Ueda, et al. 2008). However, studies also conclude that agonists may be designed to interact in combination with antagonists either to desensitize the CXCR4 receptor or to induce cytotoxic T-cell response from the immune system, leading to cancer cell death (Peggs, et al. 2009; Adil, et al. 2016). The current literature shows that peptide mimics, based on the backbone of SDF-1, are the only molecules designed so far that exhibit agonist activity at the CXCR4 receptor (Vlieghe, et al. 2010). Despite this success, the discovery of small molecule agonists that do not rely on the SDF-1 peptide backbone may be more effective and generally more versatile due to increased bioavailability, pharmacokinetic activity, membrane permeability, and metabolic stability (Craik, et al. 2013; Vlieghe, et al. 2010). For these reasons, a series of new small molecule drugs was synthesized in a lab located on the Northwestern Evanston, IL campus by Dr. Gary Schiltz based on an in silico docking procedure utilizing CXCR4 X-ray crystallographic information (Fig. 1; Mishra, et al. 2016). The entire series was initially screened to test for agonist and antagonist behavior. Of the series, NUCC-54118 is a prototypical member and showed the most robust agonist activity upon initial screening. This compound was therefore of the greatest interest in the series as more potent activity in vitro suggests stronger effects in vivo. Such effects range from stem cell migration to cancer cell death, highlighting the importance of 54118 as a therapeutic if it were to be a robust agonist. This compound was tested on the aggressive human melanoma cell line C8161, gifted by Dr. Mary J. Hendrix from the Robert H. Lurie Comprehensive Cancer Center in Chicago, IL, which is known to express high levels of CXCR4 receptors. Figure 1. The structure of small molecule 54118, which is an archetypal example of a new series of synthesized structures designed to act as agonists to the CXCR4 receptor.   Methodology: Cell Culture: Human melanoma cell line C8161, which endogenously expresses CXCR4 chemokine receptors, was plated on coverslips and incubated with fluorescent Fura-2 dye in order to be visible under the microscope for calcium imaging. Calcium buffer warmed to 37° C was used to create a solution of 1 mL calcium buffer and 4 ?L Fura-2 dye per coverslip tested. Typically, four to eight coverslips were tested per round of cell plating. Under the cell culture hood, coverslips with C8161 cells were transferred to six-well plates, each well containing an arbitrary amount of buffer. Each coverslip was washed three times by adding in and vacuuming out buffer. Each well was then filled with 1 mL of the calcium buffer and Fura-2 solution for 30 minutes for cells to absorb the dye. After this step, cells were protected from direct contact with light by the six-well plate with aluminum foil. Each coverslip was then washed three times and left in buffer for 30 minutes. Calcium Imaging Assay: Calcium-ion imaging assay was used to assess the potential agonist activity of 54118, as calcium mobilization in cells is a prominent downstream effect of activation of CXCR4 receptors. Fura-2 changes its fluorescence properties when bound to calcium and, therefore, calcium imaging recorded changes in individual cell Fura-2 fluorescence levels, where greater activation of CXCR4 receptors result in greater changes in fluorescence. The results from injecting different concentrations of 54118 were used to determine their respective impacts on CXCR4 receptors. Dose response relationships were generated using Microsoft Excel to normalize and graph the data. The Olympus 1×71 Microscope and MetaFluor Fluorescence Ratio Imaging Software were used for calcium imaging. One coverslip was placed in the stage of the microscope and filled with 1 mL of 37o C calcium containing physiological buffer to maintain cells in viable condition. The perfusion system consisted of an adjustable vacuum and a buffer line that ran to the coverslip, so that tested substances could be washed away, new calcium buffer re-added, and other substances tested on the same coverslip of cells. A specific concentration of 54118 (0.2 ?M, 2.0 ?M) or 200 nM SDF-1 in aqueous solution was injected directly in the buffer above the cells. The graph on the computer showed the cells’ relative measure of intracellular calcium, in response to the added substance over time. After a change of calcium level was detected and returned to baseline, the cells were washed. AMD3100, a known antagonist of CXCR4, was injected followed by the same previous concentration of 54118. AMD3100 binds to the CXCR4 receptor and inhibits CXCR4-mediated signals. After another wash, ATP was injected to test the viability of the cells. Throughout this procedure, the injected substance and calcium buffer solution were mixed above the cells using a Novamed P1000 pipette. This helped ensure that the cells were able to fully interact with the substance and thus yield accurate results. For the four specific calcium imaging experiments shown in the results, drugs were injected and washes were completed at various times. In Fig. 2a, SDF-1 was injected at 70 seconds, washed after noting a response, and 2.0 ?M 54118 was injected at 360 seconds. In Fig. 2b, 2.0 ?M 54118 was injected at 180 seconds, then washed after noting a response; AMD3100 was injected at 540 seconds, 2.0 ?M 54118 was injected again at 630 seconds, then washed after noting a response; ATP was injected at 930 seconds. In Fig. 2c, 0.2 ?M 54118 was injected at 90 seconds, then washed after noting a response; AMD3100 was injected at 540 seconds, 0.2 ?M 54118 was injected again at 660 seconds, then washed after noting a response; ATP was injected at 900 seconds. Results: The aim of the investigation was to test if 54118 was able to activate CXCR4 receptors. Fig. 2 shows three archetypal results from 17 calcium imaging runs. Each colored line represents a single cell; the change in the ratio of Fura-2 fluorescence levels in response to certain chemical injections is displayed through time. A lack of subsequent cell response after injecting AMD3100 confirms 54118 was effectively binding to the CXCR4 receptor. A robust cell response to ATP, an agonist of ionotropic calcium permeable purinergic receptors, indicated that the cells were viable for experimentation. Fig. 2a displays the response of the melanoma cell line C8161 to 200 nM SDF-1, the CXCR4 natural ligand, and 2.0 ?M 54118. SDF-1 stimulated a 1.2 to 1.4-fold increase in Fura-2 ratios of all cells. The subsequent injection of 2.0 ?M 54118 also stimulated a 1.2 to 1.6-fold change response in calcium levels of many of the cells. Fig. 2b illustrates C8161 being tested at 2.0 ?M 54118, which caused a 1.4 to 1.6-fold increase in Fura-2 ratios of all cells. Injection of 2.0 ?M 54118 after blocker AMD3100 resulted in attenuated responses from only a few cells. Injection of ATP caused a 1.2 to 1.4-fold change response. Fig. 2c illustrates C8161 being tested at 0.2 ?M 54118, a ten-fold decrease in drug concentration compared to that tested in Figure 2b, which caused a 1.2 to 1.4-fold increase in Fura-2 ratios of all cells. Injection of 0.2 ?M 54118 after blocker AMD3100 resulted in no response from the cells. Injection of ATP caused a 1.1 to 1.3-fold change response. Figure 2. Representative examples of C8161 melanoma cell line in response to (a) Endogenous ligand SDF-1 followed by 2.0 ?M agonist drug 54118 (b) 2.0 ?M agonist drug 54118 before and after adding 10 ?M antagonist AMD3100 and (c) 0.2 ?M agonist drug 54118 before and after adding 10 ?M antagonist AMD3100. Fig. 3 shows C8161 average cell response after the injection of 200 nM SDF-1, 2.0 ?M 54118, and 0.2 ?M 54118. Error bars represent standard deviation error, and the responses from 2.0 ?M 54118 and 0.2 ?M 54118 were highly significant, as indicated by one-tailed t-tests (p < 0.001). Figure 3. Mean (standard deviation) intracellular Fura-2 ratios of C8161 melanoma cell line at the baseline and at the height of response after injection of 200 nM SDF-1, 2.0 ?M 54118, and 0.2 ?M 54118. One-tailed t-tests confirmed that responses were significant for SDF-1 and were highly significant for 2.0 ?M 54118 and 0.2 ?M 54118. One star indicates a p-value less than 0.05, and three stars indicates a p-value less than 0.001.   Discussion: The results of this investigation support the idea that the small molecule 54118 binds to the CXCR4 receptor. The melanoma cell line C8161 responded to the drug 54118 as if it were an agonist in a dose-dependent manner (Fig. 2). This was validated by comparing the average Fura-2 fluorescence level at the baseline of the run and at the height of response after 54118 was injected. One-tailed t-tests calculated p-values less than 0.001 that confirm 54118 significantly increased CXCR4 receptor activity (Fig. 3). Importantly, the effects of 54118 were blocked by the selective CXCR4 antagonist AMD3100, further indicating that 54118 acts as a selective CXCR4 agonist (Fig. 2). The calcium imaging assay is able to record the consequences of activation of the CXCR4 receptor through the mobilization of internal calcium stores located in the endoplasmic reticulum. The binding of an agonist to the receptor allows for the G-protein mediated activation of the phospholipase C beta pathway. This pathway activates the IP3 receptor on the endoplasmic reticulum (ER), leading to the release of intracellular stores of calcium from the ER where calcium is often sequestered in cells. The Fura-2 dye binds to the free calcium released into the cytoplasm and the subsequent fluorescence levels can be detected through calcium imaging (Busillo and Benovic 2007). Agonist activity on the CXCR4 receptor has been shown to activate numerous intracellular signaling pathways which induce effects ranging from stem cell migration to cancer cell metastasis (Lewellis and Knaut 2012; Ueda, et al. 2008). SDF-1 is the cognate agonist of the CXCR4 receptor and has an extensive binding pocket (Mithal, et al. 2012). However, this makes discovering a small-molecule agonist challenging because small molecules may not bind securely and consistently to precisely the same site. 54118 was therefore created based on research findings that indicated the specific residues integral for small-molecule binding and modeled using an in silico docking procedure (Fig. 1; Mysinger, et al. 2012; Mishra, et al. 2016). The present results are highly noteworthy for a number of reasons. There has been considerable interest in the pharmacology of CXCR4 receptors because of its involvement in the pathogenesis of HIV-1 infection, cancer, stem cell biology, and inflammation. This has led to the development of numerous CXCR4 antagonists. Some of these have found clinical utility, particularly for the mobilization of hematopoietic stem cells for transplant purposes (Banisadr, et al. 2011). There has also been interest in developing CXCR4 agonists for similar purposes. Such novel structures may be free of some of the side effects, including cardiotoxicity, associated with current CXCR4 ligands and may in fact act as de facto receptor antagonists in vivo owing to receptor desensitization (Bohn, et al. 2015). So far, the only CXCR4 agonists reported in the literature are based on the peptide backbone of SDF-1. However, it is doubtful that such molecules will have good bioavailability. On the other hand, the structure of the new molecule 54118 is a small molecule prototype of a new series of CXCR4 agonists which are not reliant on the structure of SDF-1. Conclusion: These experiments demonstrate that the drug 54118 acts as an agonist at the CXCR4 receptor illustrated by the fact that it causes the calcium levels of cells to significantly increase in a dose-dependent manner. 54118 is of the first family of small molecule agonists at the CXCR4 receptor, which will have greater bioavailability and fewer side effects than current CXCR4 peptide-based ligands. Knowing the effect of 54118 on the receptor may interest the study of the CXCR4 receptor internalization to gain a better understanding of the cellular interactions between the receptor and ligand, such as if 54118 activates the CXCR4 receptor upon initial binding or once internalized. Further studies may test 54118 in vivo to assess its physiological effects on the body. Prominent examples include testing 54118 and additional members of the series in the context of several disease indications, such as cancer cell metastasis and HIV-1 infectivity which both heavily utilize the CXCR4 receptor. Acknowledgements: We would like to thank our advisor Dr. Richard J. Miller for his extensive knowledge in this field of pharmacology, dedication to our work, and kindness to allow us to research in his lab. We would also like to thank our project advisor Mr. Andrew Shum, who has taught and guided us throughout this research. Special thanks to Dr. Dongjun Ren and Ms. Brittany Hopkins for always being of help. Last but not least, we would like to thank Dr. Scheppler and the rest of the Student Inquiry and Research (SIR) staff for allowing us this incredible opportunity to explore our research interests.

References

1. Adil, D., Loo, K., Pauli, M., Sanchez-Rodriguez, R., Sandoval, P., Taravati, K., …Rosenblum, M. (2016). Tumor immune profiling predicts response to anti–PD-1 therapy in human melanoma. The Journal of Clinical Investigation. doi: 10.1172/JCI87324.
2. Banisadr, G., Frederick, T.J., Freitag, C., Ren, D.J., Jung, H., Miller, S.D., & Miller, R.J. (2011). The role of CXCR4 signaling in the migration of transplanted oligodendrocyte progenitors into the cerebral white matter. Neurobiology of Disease, 44(1), 19-27. doi: 10.1016/j.nbd.2011.05.019.
3. Bohn, L., Lohse, M., Nitabach, M., Taghert, P., & Smit, M. (2015). Exploring the biology of GPCRs from in vitro to in vivo, Molecular Pharmacology, doi: 10.1124/mol.115.100750.
4. Busillo, J., & Benovic, J. (2007). Regulation of CXCR4 Signaling. Biochimica et Biophysica Acta, 1768(4), 952-963.
5. Craik, D.J., Fairlie, D.P., Liras, S., & Price, D. (2013). The future of peptide-based drugs. Chemical biology and drug design, 81(1), 136-147. doi: 10.1111/cbdd.12055.
6. Kroeze, W.K., Sassano, M.F., Huang, X.P., Lansu, K., McCorvy, J.D., Giguere, P.M., …Roth, B.L. (2015). PRESTO-TANGO: an open-source resource for interrogation of the druggable human GPCR-ome. Nature Structural and Molecular Biology, 22(5), 362-369.
7. Lewellis, S.W., & Knaut, H. (2012). Attractive guidance: How the chemokine SDF1/CXCL12 guides different cells to different locations. Seminars in Cell and Development Biology, 23(3), 333-340. doi: 10.1016/j.semcdb.2012.03.009.
8. Liang, Z., Zhan, W., Zhu, A., Yoon, Y., Lin, S., Sasaki, M., …Shim, H. (2012). Development of a unique small molecule modulator of CXCR4. PloS one 7, e34038, doi: 10.1371/journal.pone.0034038.
9. Miller, R.J., Banisadr, G., & Bhattacharyya, B.J. (2008). CXCR4 signaling in the regulation of stem cell migration and development. Journal of Neuroimmunology, 198(1-2), 31-38. doi: 10.1016/j.jneuroim.2008.04.008.
10. Mishra, R., Shum, K., Platanias, L., Miller, R., & Schiltz, G. (2016). Discovery and characterization of novel small-molecule CXCR4 receptor agonists and antagonists. Scientific Reports, 6, 30155. doi:10.1038/srep30155.
11. Mithal, D. S., Banisadr, G. & Miller, R. J. (2012). CXCL12 Signaling in the Development of the Nervous System. Journal of Neuroimmune Pharmacology: the Official Journal of the Society on NeuroImmune Pharmacology, doi: 10.1007/s11481-011-9336-x.
12. Murdoch, C. (2000). CXCR4: chemokine receptor extraordinaire [Abstract]. Immunology Reviews, 177(1), 175-184.
13. Mysinger, M., Weiss, D., Ziarek, J., Gravel, S., Doak, A., Karpiak, J., …Volkman, B. (2012). Structure-based ligand discovery for the protein-protein interface of chemokine receptor CXCR4. Proceedings of the National Academy of Sciences USA, 109(14), 5517-5522, doi: 10.1073/pnas.1120431109.
14. Peggs, K.S., Quezada, S.A., & Allison, J.P. (2009). Cancer immunotherapy: co-stimulatory agonists and co-inhibitory antagonists. Journal of Translational Immunology, 157(1), 9-19. doi: 10.1111/j.1365-2249.2009.03912.x.
15. Petit, I., Jin, D., & Rafii, S. (2007). The SDF1-CXCR4 signaling pathway: a molecular hub modulating neo-angiogenesis. Trends in Immunology, 28(7), 299-307. doi: 10.1016/j.it.2007.05.007.
16. Ueda, S., Kato, M., Inuki, S., Ohno, H., Evans, B., Wang, Z., …Fujii, N. (2008). Identification of novel non-peptide CXCR4 antagonists by ligand-based design approach. Bioorganic & Medicinal Chemistry Letters, 18(14), 4124–4129, doi: 10.1016/j.bmcl.2008.05.092.
17. Vlieghe, P., Lisowski, V., Martinez, J. & Khrestchatisky, M. (2010). Synthetic therapeutic peptides: science and market. Drug Discovery Today, 15(1-2), 40-56, doi: 10.1016/j.drudis.2009.10.009.
18. Zlotnik, A., Yoshie, O., & Nomiyama, H. (2006). The chemokine and chemokine receptor superfamilies and their molecular evolution. Genome Biology, 7(12), 243. doi: 10.1186/gb-2006-7-12-243.