Epidermal growth factor receptor I (EGFR) is overexpressed in all aggressive cancers of epithelial origin, including squamous cell head and neck (90–100%), glioma (90–100%), non-small cell lung (75–90%), colorectal (80–85%), breast (20–30%), and cervical cancers [1
]. Anti-EGFR antibodies—e.g., cetuximab, panitumumab, necitumumab, and nimotuzumab—have been used for treating EGFR-positive cancers [6
]. Except for nimotuzumab, these antibodies have all been associated with significant cutaneous toxicity in 45–100% of patients [9
]. In contrast, nimotuzumab is better tolerated [9
] and has low skin toxicities because of its “affinity optimized” binding characteristic, which ensures a low transient binding to low-EGFR-expressing healthy tissues such as the skin [14
The sensitive and quantitative properties of positron emission tomography (PET) combined with the high affinity and specificity of antibodies make them the probe of choice for imaging and radiotherapeutic applications. Radiolabeling antibodies with radiometals is often carried out by conjugating the chelator randomly to one of the lysine residues on the antibody. We and many other authors have shown that random conjugation results in a decrease in the immunoreactivity and binding and may significantly change the biodistribution of the antibody [15
]. Site-specific conjugation often preserves the properties of the antibody/probe and hence results in optimal in vivo characteristics [16
], and is particularly important when high labeling ratios are desired [17
]. Here, we report the use of the SpyTag/SpyCatcher system to site-specifically label monoclonal antibody nimotuzumab for PET imaging (using 89
Zr) and alpha particle therapy (using 225
Ac). We chose the SpyTag/Catcher system as it is efficient, stable, and requires minimal antibody modification. It can also be used in a variety of different buffers, pHs, etc. Applications of the SpyTag/SpyCatcher system have been previously reviewed [16
]. PET imaging using 89
Zr-labeled antibodies allows for patient stratification based on the expression of the molecular target. For example, 89
Zr-atezolizumab PET/CT imaging was recently used to stratify patients who will respond to anti-PD-L1 therapy [18
]. Therefore, there is great interest in 89
Zr-labeled immunoPET probes with 76 ongoing clinical trials [19
Ac-labeled peptide and antibody radiotherapeutics also have gained significant clinical interest, with 10 of these currently in clinical development [22
]. The characteristics of 225
Ac which include a t1/2
of 10.0 days, and emission of 5 αs with energy 6–8 MeV (cumulatively 28 MeV/decay) within a range of 50–80 μm with a linear energy transfer (LET) of up to 0.16 MeV/μm make it ideal for antibody/peptide targeted radiotherapy. Here, we have expressed nimotuzumab as a fusion protein with SpyTag at the C-terminus. In the first application, site-specific labeling with 89
Zr is used for microPET/CT imaging in a mouse model of EGFR-positive tumor. In the second application, nimotuzumab-SpyTag is labeled with 225
Ac for radiotherapy. In both cases, SpyCatcher conjugates are prepared and then conjugated to the antibody. Site-specific labeling does not alter the affinity and demonstrates the use of the SpyTag/SpyCatcher (Scheme 1
A,B) as a simple method for radiolabeling antibodies with different radionuclides.
This study demonstrates, to our knowledge, the first use of the SpyTag/SpyCatcher for the site-specific labeling of radionuclides to antibodies. We have explored the use of the SpyTag/SpyCatcher system in this context to identify its benefits and limitations. We performed in vitro experiments and determined that the reaction is completed in 15 min at 10 µM of antibody-SpyTag and 20 µM of ∆N-SpyCatcher-DFO. The system is robust in terms of buffer and pH and can withstand high temperatures. These characteristics make it a favorable agent for the reproducible labeling of antibodies with radionuclides. The two radionuclides/chelators 89Zr/DFO and 225Ac/macropa performed well, indicating the general utility of this method for radiolabeling.
There was no statistical difference in the reactions performed with either nimotuzumab or control IgG, indicating that the identity of the IgG did not affect the reaction and the method should be applicable to any IgG. Additionally, of note is that free ∆N-SpyCatcher-DFO was only observed when it was in molar excess, indicating that there is no need for downstream purification if the IgG-SpyTag is kept in a slight molar excess. The robustness of the ∆N-SpyCatcher to heating and pH, although not exploited in this manuscript, offers unique opportunities to use a hot-conjugation approach, where the radionuclide complex is prepared prior to the bioconjugation step. In that case, the SpyCatcher-chelator would be radiolabeled under harsh conditions and appended to the IgG-SpyTag under gentle conditions (incubating the IgG-SpyTag in PBS with the radiolabeled SpyCatcher for 15 min). Many chelators favor heat; one example of this is 89
Zr-DOTA, which has an improved stability over many other chelators including 89
Zr-DFO. However, labeling DOTA with 89
Zr requires heating at 95 °C for 1 h, a condition incompatible with IgGs [29
]. DOTA also forms a stable complex with 225
Ac, but obtaining high radiochemical yields requires heating the DOTA-conjugated antibody/peptide [30
], which was the initial motivation for developing macropa-H2 [34
]. The stability of the SpyCatcher compared to the probe being labeled opens up new radionuclides and new labeling strategies not amendable to IgGs, including direct radiolabeling approaches, which are too harsh for IgGs [24
]. The method can be extended to prepare conjugates with many repeats of the same compound—for example, to improve the specific activity or the amount of drug delivered [17
]. Alternatively, the approach can be used to prepare multimodal theranostic products without interfering with the immunoreactivity of the antibody.
A caveat of using the system to label an IgG-SpyTag is that the SpyCatcher needs to be at 20 µM for efficient labeling. In future studies, to pre-label the SpyCatcher conjugate with the radionuclide we are interested in using the second generation, which requires several-fold less (100 nM) than the one used in this study [26
]. This would be in line with the conditions normally used for radiolabeling. An advantage of this strategy is that small amounts of antibody can be efficiently radiolabeled using a pre-labeling strategy. The SpyCatcher-radionuclide can be prepared anywhere and shipped, and the required amount can be reacted on site with an IgG of interest.
As an improvement to our previous work on the site-specific labeling of antibodies with the NIR dye IRDye800CW [28
], we chose to use the ∆N-SpyCatcher, which has a reduced immunogenicity [25
]. In previous work, we used a 1.5-fold excess, reacted for 3 h, and used additional purification steps to remove unreacted SpyCatcher. Here, we showed that using a 1:1 ratio in labeling can be performed in as little as 15 min and requires no downstream purification. Consistent with our previous work with fluorescent antibodies, we observed a higher than expected uptake in the kidneys. We expected there may be some free SpyCatcher-IRDye800CW, which was one of the reasons we performed the SEC purification of the ∆N-SpyCatcher to completely remove any impurities, yet we still saw an increase in kidney uptake (Figure 4
). Given that fluorescence and DFO-conjugated antibodies are different, we now expect that this may be an issue with the maleimide linker, which has previously been shown to be unstable in vivo [35
]. This apparent instability affected the tumor uptake of 89
Zr-labeled antibody when compared with the randomly labeled probe 89
]. A significantly (p
< 0.05) lower %IA/cc was observed for the 89
Zr-nimotuzumab-SpyTag-ΔN-SpyCatcher (6% at 48 h) than for the 89
Zr-nimotuzumab probe (~12% between 24 and 72 h) for randomly labeled nimotuzumab, as previously reported [27
]. Similarly, other major organs such as the blood, kidneys, and liver showed striking differences in uptake compared with the randomly labeled nimotuzumab probe. Future work will explore more stable NHS/NCS chemistries.
In spite of the above in vivo limitations, site-specifically labeled antibody 225
Ac-nimotuzumab-SpyTag-∆N-SpyCatcher was more effective at controlling tumor growth when compared with control IgG. The rather low therapeutic efficacy of the 225
Ac-nimotuzumab-SpyTag-∆N-SpyCatcher is likely due to the low in vivo stability of the maleimide bond, which reduced the bioavailability of the construct and hence the tumor uptake. The tumor uptake of 89
Zr-nimotuzumab-SpyTag-∆N-SpyCatcher in this study was lower than 89
Zr-nimotuzumab shown previously by our group [27
]. The therapeutic effect could be enhanced by optimizing dosing, specifically by adding a third or fourth dose to the treatment regimen and decreasing the time between doses. The doses were administered 14 days apart, and 6/8 mice in the treatment group showed a delay in tumor growth observed for about 30 days, or 14 days following the last dose.
4. Experimental Section
The chemicals used in the conjugation, radiolabeling, and purification steps were American Chemical Society reagent grade or better. Water and buffers were rendered metal-free by passing them through a column of Chelex-100 resin, 200–400 mesh (Bio-Rad Laboratories, Inc., Hercules, CA, USA), and sterile-filtered through a 0.22 µm filter. The bifunctional chelating agent deferoxamine-maleimide (DFO-maleimide) was obtained from Macrocyclics (Plano, TX, USA). We synthesized macropa-SCN from macropa-NH2
obtained from Cornell University (J. J. Wilson Lab, Ithaca, NY, USA) according to a reported procedure [34
] under a material transfer agreement. Macropa-SCN was then converted to macropa-PEG6
-maleimide using a bifunctional NHS-PEG6
-maleimide, as reported earlier [15
Zr-oxalate was received from Washington University School of Medicine (St. Louise, MO, USA). 225
Ac was supplied as the 225
Ac-nitrate by the Canadian Nuclear Laboratories (CNL, Chalk River, ON, Canada) under the terms of a research collaboration.
-Cys-ΔN-SC (Figure 1
C) was purchased as a gblock from Integrated DNA Technologies. The plasmid pDEST14-SpyCatcher [23
] was purchased from Addgene and digested with Nde
1 and Sac
1 to get rid of entire His6
-SpyCatcher. The PCR-amplified gBlock of His6
-Cys-ΔN-SC was inserted into the Nde
1 and Sac
1 digested pDEST14 backbone using Gibson assembly to generate pDEST14-Cys-ΔN-SC. Previously reported plasmids for the expression of nimotuzumab-SpyTag (pFUSEss-CHIg-Nimotuzumab-hG1-SpyTag/pFUSEss-CLIg-Nimotuzumab-hG1) or control-IgG-SpyTag (pFUSEss-CHIg-Anti-MBP-hG1-SpyTag/pFUSEss-CLIg-Anti-MBP-hG1) [28
4.3. Expression and Purification of His6-Cysteine-ΔN-SpyCatcher and Antibodies
-Cys-ΔN-SpyCatcher expression plasmid was transformed into RosettaTM
(DE3) competent E. coli
cells (Novagen, Madison, WI, USA) and plated on LB agar containing carbenicillin (100 µg/mL) and chloramphenicol (34 µg/mL). A single colony was picked and cultured overnight in LB with carbenicillin (100 µg/mL) and chloramphenicol (34 µg/mL) at 37 °C. The overnight culture was diluted 100-fold and grown at 30 °C until the OD600
reached 0.8, then induced with 0.25 mM IPTG and grown at 30 °C for 4 h. Sample was lysed and purified on a GE Healthcare AKTA FPLC system using a Nickel affinity column (GE healthcare). Nimotuzumab-SpyTag and control-IgG-SpyTag were expressed using the Gibco™ Expi293™ Expression System (Life Technologies, Carlsbad, CA, USA, Catalog Number: A14635) and purified as described previously [28
4.4. Conjugation of Chelator to Cys-∆N-SpyCatcher
Cys-∆N-SpyCatcher (Scheme 1
A,B) was first reduced with a 50-fold molar excess of tris(2-carboxyethyl)phosphine (TCEP; Sigma, St. Louis, MO, USA, # C4706) for 2 h at room temperature to form SH-∆N-SpyCatcher (Scheme 1
A,B). To this mixture, a 15-fold molar excess of DFO-maleimide or macropa-PEG6
-maleimide was added and allowed to react overnight at 4 ºC. Excess chelator and TCEP were removed by passing the solution through a 5 mL zeba spin desalting columns (7K MWCO; Fisher Scientific, Waltham, MA, USA). ∆N-SpyCatcher-DFO or ∆N-SpyCatcher-macropa was buffer exchanged to PBS and concentrated with Amicon Ultra centrifugal filters (3 kDa MWCO; Millipore, Burlington, MA, USA). The chelator-labeled SpyCatcher was stored at −80 °C until further use.
4.5. Bioconjugation of IgG-SpyTag with Chelator-∆N-SpyCatcher
Unless otherwise indicated, a 1:1 molar ratio of SpyTag to SpyCatcher was used. Nimotuzumab-SpyTag and control IgG-SpyTag (10 µM) were ligated to pre-labeled ∆N-SpyCatcher-DFO or ∆N-SpyCatcher-macropa (20 µM) for 1 h at room temperature in PBS pH 7.2 and terminally sterilized by passing through a 0.22 µm filter (Ultrafree MC; Millipore, Burlington, MA, USA).
4.6. Quality Control of Immunoconjugates
The purity of the constructs was determined by HPLC analysis or a bioanalyzer. HPLC was performed using an XBridge Protein size exclusion column (200 Aº 3.5 µM; Waters, Milford, MA, USA) and eluted with 1 × PBS buffer, pH 7.0 at a flow rate of 0.6 mL/min. Proteins were monitored at 280 nm. Molecular weight and purity were characterized by electronic electrophoresis (2100 Bioanalyzer system, Agilent, Santa Clara, CA, USA). The molecular weight (MW) and purity were measured using a high-sensitivity protein 250 Kit (Agilent, Santa Clara, CA, USA, cat # 5067-1575), according to the manufacturer’s protocol. The molecular weight and peak areas were calculated using the 2100 Expert software (Agilent, Santa Clara, CA, USA).
4.7. Cell Culture
The human cancer cell line MDA-MB-468 that expresses EGFR and the negative control cell line (no EGFR expression) MDA-MB-435 were obtained from ATCC (Rockville, MD, USA). Cells were propagated by serial passage in 90% DMEM and 90% RPMI medium, supplemented with 10% fetal bovine serum (FBS, Sigma-Aldrich, St. Louis, MO, USA). Cells were grown in a humidified atmosphere with 5% CO2 at 37 °C. Prior to in vitro and in vivo use, the cells were detached using trypsin-EDTA.
4.8. Flow Cytometry
In vitro binding studies were carried out in EGFR-positive cancer cells MDA-MB-468. 1 × 105 cells were collected and washed with 1 × PBS + 2% FBS. Antibodies were titrated at a minimum of a 10-fold molar excess onto the cells at concentrations between 2000–0.03 nM in a 12-point curve, incubated for 30 min at room temperature followed by 15 min on ice. Cells were washed and suspended in a 1:50 dilution of FITC-labeled Goat F(ab’)2 fragment anti-human IgG (H + L) antibody (Beckman Coulter, Brea, CA, USA) and incubated for 30 min on ice in the dark. Cells were washed and suspended in 1 × PBS + 2% FBS and analyzed using a Gallios flow cytometer (Beckman Coulter, Brea, CA, USA) on the FL1 channel. FlowJo V.10.0.8 was used for analysis and to determine the mean fluorescence intensity. For the fitting and normalization of the mean fluorescence intensity, the top was globally fit and the bottom was fit to the average and then normalized. The KD was calculated based on a non-linear regression curve using GraphPad Prism 6.
4.9. Radiolabeling with 89Zr and 225Ac
Radiolabeling with 89
Zr was performed as described previously [27
]. The purity of the radiolabeled immunoconjugates was determined using size exclusion radio-HPLC and iTLC. Radioactivity was detected using a flow-through radio-HPLC-detector (Flow-RAM, Broomhill, UK). The final solution was formulated in PBS. A radiochemical purity (RCP) of more than 95% was considered sufficient for in vitro and in vivo experiments.
In a typical 1-step procedure, 225
Ac-nitrate (1.0 MBq) dissolved in 0.05 M of HCl (Optima grade, Fisher scientific, Waltham, MA, USA) was added to a 1.5 mL microtube, and the activity was determined using a dose calibrator. To this, 150 mM of ammonium acetate (pH 6.0, 100 µL) and nimotuzumab-SpyTag-∆N-SpyCatcher-macropa or macropa-control-IgG (833 µg) were added. The pH of the reaction was determined by spotting 1 µL of the reaction mixture onto Hydrion pH paper (range, 5.0–9.0) (Sigma-Aldrich, St. Louis, MO, USA); the pH of a typical reaction was 5.8. The reaction mixture was incubated at 37 ºC on a shaker at 650 RPM for 50 min [37
]. After this, a small aliquot (0.5 µL) was spotted on a strip of instant thin-layer chromatography silica gel impregnated paper (iTLC-SG, Agilent Technologies, Santa Clara, CA, USA) to determine the extent of the incorporation of actinium onto the protein using a mobile phase of 50 mM of sodium citrate (pH 5.2). The purification of 225
Ac-labeled tracer was conducted using Amicon Ultra-4 centrifugal filters (10 K, EMD Millipore, Burlington, MA, USA) with PBS.
4.10. Stability of Radioimmunoconjugates
The stability of the radioimmunoconjugates was evaluated in vitro in saline and human plasma. A total of 50 µL of radiolabeled compound was added to 1 mL of human plasma or saline to make a final concentration of 20 MBq/mL (89
Zr-radiolabeled conjugate) or 1 MBq/mL (225
Ac-radiolabeled conjugate), followed by incubation at 37 °C for 7 to 10 days (n = 3). Aliquots were taken at different time points and analyzed for radiochemical purity using iTLC [38
4.11. Biodistribution, Tumor Xenografts and MicroPET/CT Imaging
All the animal studies were approved by the University of Saskatchewan Animal Care and Use Committee in accordance with the guidelines set forth on Use of Laboratory Animals (protocol # 20170084). Female athymic CD-1 nude mice were purchased from the Charles River Laboratory (Sherbrooke, QC). The mice were housed under standard conditions in approved facilities with 12 h light/dark cycles and given food and water ad libitum throughout the duration of the studies. For inoculation, MDA-MB-468 or MDA-MB-435 cells were suspended at 5 × 107 cells/mL in a 1:1 mixture of media without FBS: Matrigel (BD Biosciences, Mississauga, ON, Cananda). Each mouse was injected in the right flank with 0.1 mL of the cell suspension.
The biodistribution of 89Zr-nimotuzumab-SpyTag-∆N-SpyCatcher and 89Zr-control-IgG-SpyTag-∆N-SpyCatcher. was performed in normal female athymic CD-1 nude mice following a tail vein injection of 8–12 MBq (specific activity, 0.5 MBq/µg) of the probes. Mice were sacrificed at 24 and 144 h post injection, and the activity in organs was measured using a gamma counter (Wallac Wizard 1480, PerkinElmer, Waltham, MA, USA) and expressed as the % injected activity per gram (%IA/g).
Mice (n = 4/group) were used for imaging studies when the tumors reached approximately 200–400 mm3. Mice bearing MDA-MB-468 or MDA-MB-435 xenografts received an intravenous injection of 8–12 MBq (specific activity, 0.5 MBq/µg) of 89Zr-nimotuzumab-SpyTag-∆N-SpyCatcher. At 24, 48, 120, and 144 h post injection, PET and CT images were acquired using a Vector4CT scanner (MILabs, Utrecht). Mice were anesthetized using a mixture of isoflurane/oxygen (5% of isoflurane in oxygen), and in vivo whole-body PET/CT images were obtained while anesthesia was maintained (2% of isoflurane). Body temperature, heart rate, and breathing frequency were monitored continuously and kept at normal physiologic values.
The PET scans were acquired in a list-mode data format with a high-energy ultra-high resolution (HE-UHR-1.0 mm) mouse/rat pinhole collimator. Corresponding CT scans were acquired with a tube setting of 50 kV and 480 μA. PET image reconstruction was carried out with a pixel-based order-subset expectation maximization (POS-EM) algorithm that included resolution recovery and compensation for distance-dependent pinhole sensitivity, and was registered on CT and quantified using the PMOD 3.8 software (PMOD, Switzerland).
4.12. In Vitro Cytotoxicity
value of 225
Ac-nimotuzumab-SpyTag-∆N-SpyCatcher or 225
Ac-control-SpyTag-∆N-SpyCatcher immunoconjugates was determined using an IncuCyte S3 Live cell imager (Essen BioScience, Ann Arbor, MI). Briefly, 3000–5000 (MDA-MB-468) cells were seeded 24 h prior to treatment in 96-well plates. The next day, the media was removed and washed with PBS. Cells were incubated with IncuCyte®
Cytotox Red reagent diluted in complete media (1×, Essen Bioscience, Ann Arbor, MI, USA, Cat #4632) for 3 h before treatment. The cells were treated with different concentrations (3.7–0.028 kBq) of 225
Ac-nimotuzumab-SpyTag-∆N-SpyCatcher or 225
Ac-control-IgG-SpyTag-∆N-SpyCatcher and the plate was incubated at 37 °C for 30 min prior to imaging. Live cell images were captured every 2 h using a 10 × objective lens using phase contrast and fluorescence channel. During each scanning, five images were acquired until the end of the experiment (48 h). All the cell images were processed and analyzed using the IncuCyte S3 software. The red fluorescent values were generated and the EC50
values for individual compounds were calculated using GraphPad prism 6. The EC50
concentration were calculated with reference to a control sample, which represents the concentration that results in a 50% decrease in the cell number/growth/proliferation after 48 h incubation in the presence of an antibody construct [15
4.13. 225Ac Radioimmunotherapy
When xenografts averaged 50–100 mm3 in volume, the mice were randomized into 4 groups (n = 8 per group). Each group received PBS or two doses of nimotuzumab, 225Ac-nimotuzumab-SpyTag-∆N-SpyCatcher, or 225Ac-control-IgG-SpyTag-∆N-SpyCatcher via a tail vein on day 0 and 14. The tumor growth was monitored by measuring the greatest length and greatest width of each tumor using digital caliper. Then, the tumor volume was calculated using the formula tumor volume = length × width2/2. The study was terminated when the xenograft reached a volume ≥2000 mm3, and this was used to determine survival in the different groups using Kaplan–Meier curves. The individual body weights of mice were recorded during the quarantine and experimental period (every other day).
4.14. Statistical Analysis
Unless otherwise stated, all data were expressed as the mean ± SD or SEM of at least 3 independent experiments. Statistical comparisons between the experimental groups were performed either via Student t tests with Welch correction (2-group comparison) or a 1-way ANOVA with Bonferoni multiple comparison post hoc test (multiple-group comparison). Graphs were prepared and p values calculated by using GraphPad Prism (version 6; GraphPad, La Jolla, CA, USA). p values of less than 0.5 were considered significant.