Subscriber access provided by Macquarie University
Article
Affinity-Guided Conjugation to Antibodies for use in Positron Emission Tomography Mikkel Bach Skovsgaard, Troels Elmer Jeppesen, Michael Rosholm Mortensen, Carsten Haagen Nielsen, Jacob Madsen, Andreas Kjær, and Kurt Vesterager Gothelf Bioconjugate Chem., Just Accepted Manuscript • DOI: 10.1021/acs.bioconjchem.9b00013 • Publication Date (Web): 26 Feb 2019 Downloaded from http://pubs.acs.org on February 27, 2019
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 9 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Bioconjugate Chemistry
Mikkel B. Skovsgaard,† Troels E. Jeppesen, ‡ Michael R. Mortensen,† Carsten H. Nielsen, ‡ Jacob Madsen, ‡ Andreas Kjaer‡ and Kurt V. Gothelf†* †iNANO
and Department of Chemistry, Aarhus University, Gustav Wieds Vej 14, 8000 Aarhus, Denmark of Clinical Physiology, Nuclear Medicine & PET and Cluster for Molecular Imaging, Department of Biomedical Sciences, Rigshospitalet and University of Copenhagen, Blegdamsvej 9, 2100 Copenhagen N, Denmark ‡Department
Abstract: The radionuclide copper-64 is widely used in combination with biomolecules, such as antibodies, for positron emission tomography (PET). Copper-64 is ideal for the imaging of biomolecules with long circulation times due to its relatively long half-life and when conjugated to an antibody, specific cells can be targeted in vivo. Here, we have prepared a trastuzumab-chelator conjugate by using affinity-guided conjugation, in which an azide was attached to the antibody prior to a strain promoted azide-alkyne cycloaddition reaction with DBCO-PEG4-NOTA. The conjugate was benchmarked against a standard non-specific labeled trastuzumab-NOTA conjugate. The conjugates were tested for incorporation of copper-64, stability in buffer and plasma, and for tumor targeting in vivo using PET imaging of mice with xenograft tumors expressing HER2. Both conjugates showed good incorporation of copper-64 and a high stability with less than 10% degradation after 36 hours. Furthermore, both conjugates showed accumulation at the tumor site with mean uptake of 7.2 ± 2.4 %ID/g and 5.2 ± 1.3 %ID/g after 40 hours for the affinity-guided labeled trastuzumab and the non-specific labeled trastuzumab, respectively.
Antibody conjugates have a variety of clinical applications of which the use as targeting agent for positron emission tomography (PET) in diagnostics is an important example.1–3 Copper-64 (64Cu, half-life 12.7 h, 17.4% decay by β+ 0.656 MeV) is a commonly used radionuclide for PET of long circulating proteins, since the half-life is relatively long. Most often, radioactive cationic metals are conjugated to a protein through binding to a chelator already conjugated to the antibody. To obtain a high kinetic stability of the metal-chelator complex, a polydentate chelator, such as DOTA or NOTA, is required. They are both polydentate, cyclic chelators commercially available and they have been used in combination with a variety of biomolecules.4,5 Furthermore, both DOTA and NOTA have been investigated in the clinic.6,7 Several approaches to antibody conjugation have been investigated. Commonly, the methods are based on nonspecific labeling of the antibody by reaction of lysine residues on the protein with electrophiles such as NHS esters or isothiocyanate groups.4,8,9 Such conjugations yield a heterogenous population of antibody conjugates since the reaction provides little control of the positioning and gives a distribution of the number of labels per antibody. In another approach, a specific handle is introduced to a recombinant antibody by genetic engineering.10,11 This provides a high degree of control of the conjugation reaction but requires optimization for each new protein of interest. A third approach for protein conjugation relies on affinityguided conjugation to direct the modification to a specific
area on the protein.12–14 This method offers higher control over the positioning of the modification compared to nonspecific conjugation, without the need for genetic manipulation of the protein. Recently, our laboratory developed small molecule probes for the affinity-guided introduction of handles to lysine residues by interaction with a metalbinding site on the protein of interest including IgG1 antibodies.15 The conjugation reaction is performed through reductive amination between a lysine residue and an aldehyde on the probe. The chelation between the small molecule probe, metal and the protein metal-binding site increases the local concentration of the reactive aldehyde at the protein surface, while at the same time allowing the protein and probe to be at a low concentration, which limits unspecific labeling. Affinity-guided conjugation methods result in less heterogeneous protein conjugates and the risk of hampering the binding of the ligand, by reaction to or near the ligand binding site, is decreased compared to non-specific labeling. We compared the uptake of affinity-guided copper-64 conjugates in mice with the uptake of a copper-64 conjugate prepared by a commonly used nonspecific labeling method.
Affinity-guided conjugation. In this study, we have applied the antibody trastuzumab, a human IgG1 antibody against human epidermal growth factor receptor 2 (HER2), which are overexpressed in some breast cancers.16 The application of trastuzumab-chelator conjugates has previously been applied in PET diagnostics.2,4 In this study, trastuzumab was modified with NOTA through affinity-
ACS Paragon Plus Environment
1
Bioconjugate Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 2 of 9
Figure 1. Schematic illustration of the conjugation method. Antibody chelator conjugates are prepared by two different methods, affinity-guided conjugation resulting in Tz-Az-NOTA and non-specific conjugation giving Tz-SCN-NOTA. Both conjugates are subsequently labeled with copper-64. These radioactive conjugates were injected in mice, for in vivo imaging. a) Tz (2 µM), small molecule probe (20 µM), Cu(NO3)2 (60 µM), NaBH3CN (50 mM), HEPES (25 mM, pH 7.5, with 150 mM NaCl), incubated overnight at room temperature. b) Addition of EDTA (2.5 mM) followed by Amicon filtration. c) DBCO-PEG4-NOTA (10 eq), phosphate buffer (25 mM, pH 7.5). d) Tz (50 µM), isothiocyanate-NOTA (250 µM), EPPS buffer (50 mM, pH 8.8), incubated overnight at room temperature. e) Washed using Amicon filtration.
guided conjugation of an azide containing small molecule probe followed by click reaction with DBCO-PEG4-NOTA (Figure 1).15 The NOTA chelator incorporates copper-64 at room temperature and shows high stability in vivo.17 The affinity-guided conjugation method is based on small molecule probes with two metal chelating moieties, nitrilotriacetic acid (NTA) groups. These moieties can coordinate a copper ion and subsequently bind to a metal-binding site on the constant domain of the antibody, as illustrated in Figure 1. After reductive amination, the copper is removed by ethylenediaminetetraacetic acid (EDTA). An azide functionality on the probe is then used for the strain promoted azide-alkyne cycloaddition (SPAAC) reaction with the DBCO-PEG4-NOTA. The conjugation is performed with 10 eq of the probe and excess reagent as well as EDTA-Cu are removed after conjugation by ultrafiltration. Non-specific conjugation. To compare our conjugate with a non-specifically labeled conjugate, we produced NOTA-conjugated trastuzumab (Tz-SCN-NOTA) by reaction with 5 eq of isothiocyanate-NOTA (Figure 1). Radiolabeling. The conjugates are labeled with copper64 (shown for Tz-Az and Tz-Az-NOTA in Figure 2A) by addition of an aliquot of [64Cu]CuCl2 with 400-600 MBq to the conjugates in 0.1 M NH4OAc. After 15 min incubation at 25 °C, full incorporation of copper-64 is observed as an excess of antibody conjugate is used. As the NTA chelators
on the small molecule probe also bind copper, a control experiment is performed to ensure that EDTA removes the copper bound to the NTA groups. The NTA groups bind copper less strongly than NOTA. In vivo, this could result in the release of the nuclide from the conjugate if copper is not removed beforehand.18–20 The copper-64 labeling was performed on both the Tz-Az-NOTA conjugate as well as the Tz-Az conjugate before both samples were incubated with EDTA for 15 min. Using size exclusion chromatography (SEC), the release of copper-64 from the NTA groups by EDTA was investigated. A signal just after 6 mins in both the UV and radio chromatograms is observed, corresponding to the [64Cu]Cu-conjugate (Figure 2B and Supporting Figure S1). The analysis indicates full release of copper-64 from the NTA groups upon EDTA treatment, while copper-64 remains chelated by NOTA (Figures 2C and 2D). Furthermore, Tz-Az-NOTA shows higher incorporation of copper-64 than Tz-Az (Figure 2E). Number of accessible chelators. We investigated the number of accessible chelators on the conjugates by titration with non-radioactive copper doped with copper-64, which was analyzed by TLC (Supporting Figure S2). The analysis indicates an accessible chelator to antibody ratio of 0.94±0.12 for the affinity modified conjugates and 3.43±0.50 for the non-specifically labeled conjugate, reflecting the number of ligands per antibody.
ACS Paragon Plus Environment
2
Page 3 of 9 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Bioconjugate Chemistry Purification of the radioactive conjugate. After incorporation of copper-64 and addition of EDTA, the conjugates were purified using PD-10 columns pre-eluded with BSA. The conjugates were eluted with PBS buffer and the
fractions with the highest radioactive were combined. The purified conjugates showed a high radiochemical purity of >99%, assessed by HPLC (Figure 3). The conjugates, [64Cu]Cu-Tz-Az-NOTA (n=3) and [64Cu]Cu-Tz-SCN-NOTA
Figure 2. (A) Schematic illustration of the incorporation of copper-64 in Tz-Az and Tz-Az-NOTA followed by treatment with 2 mM EDTA. Tz-Az is shown in the upper panel and Tz-Az-NOTA in the lower panel. (B-D) UV and radio chromatograms for the copper64 labeled conjugates. (B) Analysis of [64Cu]Cu-Tz-Az-NOTA before addition of EDTA. (C) Analysis of [64Cu]Cu-Tz-Az after addition of EDTA. (D) Analysis of [64Cu]Cu-Tz-Az-NOTA after addition of EDTA. (E) Copper-64 incorporation with or without EDTA addition in Tz-Az and Tz-Az-NOTA. Copper-64 is stripped from the NTA groups showing large decrease in protein-bound radioactivity for [64Cu]Cu-Tz-Az and minor decrease in protein-bound radioactivity for [64Cu]Cu-Tz-Az-NOTA.
ACS Paragon Plus Environment
3
Bioconjugate Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 4 of 9
Figure 3. Analysis of the purified copper-64 conjugates using size exclusion chromatography. Upper panels show UV absorbance of the protein while the lower panel shows detection of radioactivity. The chromatograms show a high purity of the conjugates.
(n=2), were prepared with a radiochemical yield of 66.9 ± 6.7% and 81.4 ± 9.0%, respectively. For the animal studies, an aliquot of [64Cu]CuCl2 with approximately 600 MBq, was incubated with 4 nmol of conjugate yielding a molar activity of 132 GBq/µmol and 75 GBq/µmol for [64Cu]Cu-Tz-AzNOTA and [64Cu]Cu-Tz-SCN-NOTA, respectively, after purification. Buffer and plasma stability. To test the stability in buffer of [64Cu]Cu-Tz-Az-NOTA and [64Cu]Cu-Tz-SCNNOTA, the conjugates were incubated in PBS buffer at room temperature. The samples were analyzed by HPLC and showed no degradation of the conjugates (Supporting Figures S3 and S4). The stability of the conjugates was further tested in a plasma stability experiment, where they were incubated in 1:1 PBS buffer and mouse plasma at 37 °C for 4, 16, and 36 hours. The samples were diluted 100 times with PBS buffer at the given time points and frozen until analyzed by SDS-PAGE under reducing conditions (Supporting Figure S5). The amount of radioactivity in the bands corresponding to the antibody heavy and light chains were compared to the radioactivity in the remainder of the lane for the three time points. The experiment was made with triplicate samples and showed high stability of the conjugates in plasma, as expected (Figure 4). For [64Cu]Cu-Tz-Az-NOTA, a higher signal was observed for the heavy chain compared to the light chain, while [ 64Cu]CuTz-SCN-NOTA showed more equal conjugation to both chains, indicating a different conjugation pattern for the two conjugates. Binding affinity. The binding affinity of the conjugates was analyzed by incubation with HER2 expressing cells. The amount of binding was analyzed at different conjugate concentrations. The amount of unspecific binding to the cells was investigated by addition of 200-fold excess of unmodified trastuzumab (Supporting Figure S6). The binding affinity was calculated based on a non-linear regression model that compensates for the non-specific binding. This resulted in a better binding affinity for the [64Cu]CuTz-Az-NOTA conjugate 1.7 ± 0.4 nM compared to 3.3 ± 0.4 nM for [64Cu]Cu-Tz-SCN-NOTA. On the other hand, [64Cu]Cu-Tz-Az-NOTA also showed higher unspecific binding compared to [64Cu]Cu-Tz-SCN-NOTA, which could be
caused by the larger modification on the [64Cu]Cu-Tz-AzNOTA. Longitudinal PET imaging and biodistribution. The distribution and tumor localization of the conjugates in mice with subcutaneous human ovarian xenograft tumors, SK-OV-3, expressing HER2 were assessed by longitudinal PET/CT imaging. The mice were scanned at 16 and 40 hours after injection of the radioactive conjugates, 8.3 ± 0.1 MBq for [64Cu]Cu-Tz-Az-NOTA and 8.1 ± 0.1 MBq for [64Cu]Cu-Tz-SCN-NOTA. Furthermore, analysis of the ex vivo biodistribution was performed at 40 hours. Representative coronal and axial images for two mice injected with either [64Cu]Cu-Tz-Az-NOTA or [64Cu]Cu-Tz-SCNNOTA show localization of the conjugates at the tumor (Figure 5A). The combined results (n=5) for the 16 hours and 40 hours scans are illustrated in Figures 5B and 5C, respectively.
Figure 4. The plasma stability of the conjugates based on SDSPAGE analysis. The conjugates were incubated in a 1:1 mixture of mouse plasma and PBS at 37 °C. The samples were diluted 100x at the different time-points and frozen until analyzed by SDS-PAGE. The amount of radioactivity in the antibody bands were compared to the radioactivity in the rest of the lanes and normalized to the 4-hour sample. The experiments were performed in triplicates, and the error bars represent standard deviations.
ACS Paragon Plus Environment
4
Page 5 of 9 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Bioconjugate Chemistry
Figure 5. In vivo testing of the conjugates. (A) Longitudinal PET/CT imaging of [ 64Cu]Cu-Tz-Az-NOTA and [64Cu]Cu-Tz-SCN-NOTA in mice bearing human ovarian xenograft SK-OV-3 tumors at 16 and 40 hours after injection. (B) Quantitative analysis of in vivo uptake at 16 hours after injection (n=5). (C) Quantitative analysis of in vivo uptake at 40 hours after injection (n=4 for [64Cu]Cu-TzAz-NOTA. n=5 for [64Cu]Cu-Tz-SCN-NOTA). (D) Ex vivo biodistribution at 40 hours after injection (n=5). Error bars represents SEM. Two-way ANOVA was used to calculate significant differences (* denotes P