^99mTc-bioorthogonal click chemistry reagent for in vivo pretargeted imaging

Abstract

Metal-free click chemistry has become an important tool for pretargeted approaches in the molecular imaging field. The application of bioorthogonal click chemistry between a pretargeted trans-cyclooctene (TCO) derivatized monoclonal antibody (mAb) and a ^99mTc-modified 1,2,4,5-tetrazine for tumor imaging was examined in vitro and in vivo. The HYNIC tetrazine compound was synthesized and structurally characterized, confirming its identity. Radiolabeling studies demonstrated that the HYNIC tetrazine was labeled with ^99mTc at an efficiency of >95% and was radiochemically stable. ^99mTc-HYNIC tetrazine reacted with the TCO-CC49 mAb in vitro demonstrating its selective reactivity. In vivo biodistribution studies revealed non-specific liver and GI uptake due to the hydrophobic property of the compound, however pretargeted SPECT imaging studies demonstrated tumor visualization confirming the success of the cycloaddition reaction in vivo. These results demonstrated the potential of ^99mTc- HYNIC-tetrazine for tumor imaging with pretargeted mAbs.

Introduction

Receptor targeting radiopharmaceuticals is an area of intense research for molecular imaging and cancer therapy. Considerable efforts have been focused on the development of molecules and biomolecules with highly selective recognition and affinity for specific targets, as well as on the advancement of labeling strategies that lead to improved radiolabeling yields without compromising target affinity and recognition [1–5]. Monoclonal antibodies (mAbs) have been used for many years to deliver radionuclides to targeted tissues; however, their slow pharmacokinetics and tumor accumulation limits applications to radionuclides of moderate and long half-lives [6–7]. Extended circulation times of radiolabeled antibodies allow enough time for tumor accumulation, however it increases the dose to non-target tissues due to slow clearance [5,8–9]. Alternatively, a pretargeted approach is a strategy that attempts to overcome problems related to the slow pharmacokinetics of radiolabeled antibodies [8–10]. The pretargeting labeling strategy is a multi-step process that takes advantage of the high affinity of multidentate large biomolecules and antibodies, without the drawbacks associated with slow clearance (Scheme 1A). For example, the first step of a pretargeted mAb approach includes administration of a tagged unlabeled antibody and subsequent tumor targeting and clearance of unbound material. The second step involves injection of a radiolabeled small molecule that has high specificity and bioorthogonal reactivity for the tag on the tumor localized antibody [11–13]. The rapid in vivo pharmacokinetics of the radiolabeled small molecule are designed to maximize the accumulation of the radionuclide in tumor and decrease radioactivity in other non target tissues, resulting in superior target/non target ratios compared to directly radiolabeled antibodies and reducing patient dose [14,15]. Bioorthogonal click chemistry has become an important tool for the pretargeted approach [16,17]. In this sense, the inverse electron demand Diels-Alder reaction between a highly reactive trans-cyclooctene (TCO) and an electron deficient 1,2,4,5-tetrazine (TZ) is particularly exciting for imaging of large molecules and nanoparticles in vivo [16–18]. This strategy opens a variety of applications that could be applied to in vivo imaging and therapy. Several bio-orthogonal TZ-based cycloaddition ligations have been used for in vivo radiolabeling of biological molecules as antibodies [19,20]. The synthesis of different TZ derivates allowed radiolabeling with ¹¹¹In, ¹⁷⁷Lu, ⁶⁸Ga and subsequent use in pretargeted approaches for in vivo imaging and therapy with antibodies and polymers [21–24].

Schematic 1.

A pretargeted approach using bioorthogonal click chemistry for tumor imaging. A) A schematic of the pretargeted approach using a TCO conjugated mAb and radiolabeled TZ to target and image a tumor in vivo. B) Synthesis of the HYNIC-TZ derivative (1). C) The radiolabeling procedure for HYNIC-TZ with ^99mTc.

Technetium-99m (^99mTc) is the most widely used radionuclide for diagnosis in nuclear medicine. It is easily obtained from ⁹⁹Mo-^99mTc generator and presents a gamma emission of 140 keV and has a 6-hour half-life. These properties are suitable for routine clinical imaging, mainly in those countries where PET technology is still being established [25–27]. Several labeling methods can be used for preparing [^99mTc]-labeled biomolecules [28]. The 6-hydrazinonicotinc acid (HYNIC) is a commonly used bifunctional chelating agent [29]. It is attractive for preparing [^99mTc]-labeled peptides and proteins with high efficiency, under mild conditions, and is highly stable in vivo [30]. However, one of the main problems of HYNIC is the highly nucleophilic hydrazine group that can undergo unwanted side reactions with traces of electrophiles, such as aldehydes and ketones, present as contaminants and induce the inactivation of HYNIC [30–32]. Many different protecting groups are used to avoid HYNIC instability [31–36]. In this context, tert-butoxycarbonyl group (Boc) is widely used as a protecting group during the conjugation of HYNIC to peptides, however the deprotection step required before labeling is not compatible with protein integrity [37–39]. Recently, Surfraz et al. reported the trifluoroacetyl group (Tfa) for hydrazine-HYNIC protection that could be deprotected during ^99mTc labeling conditions on a conjugated peptide model [40,41]. They also showed that HYNIC-Tfa deprotection is promoted by a technetium complex intermediate during reduction of Tc (VII) in presence of SnCl₂ and tricine [41,42].

Several approaches are under investigation for the development of a ^99mTc radiolabeled TZ for SPECT imaging with pretargeted mAbs, phage and nanoparticles [43–47]. Development of a pretargeted/bioorthogonal approach using the TCO-TZ combination for labeling with ^99mTc would expand the application and utility of ^99mTc-labeled antibodies. Therefore, herein we present the evaluation and the potential use of bioorthogonal click chemistry applied to ^99mTc labeling of biomolecules. For this purpose, we describe the synthesis of a HYNIC-Tfa protected TZ derivative, the ^99mTc labeling process, evaluation in in vitro binding assays and in vivo SPECT imaging with pretargeted TCO conjugated CC49 mAb.

Results and Discussion

Chemistry

The application of inverse electron demand Diels-Alder cycloaddition for bioorthogonal in vivo pretargeted imaging with ^99mTc is presented in Scheme 1A. A trifluoroacetyl protected HYNIC-tetrazine derivative (1, HYNIC-TZ, Scheme 1B) was synthesized from 4-(6-methyl-1,2,4,5-tetrazin-3-yl)phenylmethanamine and a succinimidyl protected HYNIC precursor. Compound (1) was obtained in a yield of 27 %. It was stable for at least 24 h in PBS and BSA (Supplementary material, Figure S2).

Different ^99mTc labeling conditions, of temperature and reaction times, were studied when tricine was used as coligand (Scheme 1C, Table 1). We observed a HYNIC-TZ labeling yield superior to 90 % in all the cases, leading to an optimized condition when the reaction was performed at 37 °C for 15 min. The radiochemical purity of the complex (RCP) was established by RP-HPLC (Figure 1). The labeled complex was shown to be stable up to 24 h in the reaction mixture and PBS when it was incubated at 25 °C (Figure 2).

Table 1.

The effect of temperature and incubation time on labeling efficiency of tri-^99mTc-HYNIC-TZ using tricine as coligand (n=4).

temperature (°C)	reaction time (min)	% RCPa
25	60	97 ± 2
25	30	92 ± 1
37	30	97 ± 2
37	15	96 ± 1

^a% RCP: radiochemical purity expressed as percentage.

Figure 1.

RP-HPLC analysis of the radiolabeling mixture showing the unlabeled HYNIC-TZ (1) (t_R = 17.3 min) in the UV (Abs 280 nm) profile (upper) and the radiolabeled complex with tricine as coligand (tri-^99mTc-HYNIC-TZ) (t_R = 15.3 min) in the radioactive profile (lower).

Figure 2.

Radiochemical stability assay of the tri-^99mTc-HYNIC-TZ complex in the labeling reaction mixture and in PBS at 25 °C over 24 h.

To evaluate the effect of the coligands on the lipophilicity property of the ^99mTc-HYNIC-tetrazine complexes, the labeling of (1) using tricine, ethylenediaminediacetic acid (EDDA), or a mixture of tricine and EDDA as coligand was performed (see experimental conditions in Supplementary material). The n-octanol/water partition coefficients for the different ^99mTc-HYNIC-tetrazine complexes showed that the coligand tricine resulted in the least lipophilic entity with a Log P value of −0.54 ± 0.07 (Table 2). Based on these results, the in vivo behavior of the tricine complex of ^99mTc-HYNIC-tetrazine (tri-^99mTc-HYNIC-TZ) was evaluated in vitro and in vivo.

Table 2.

The lipophilicity dependence of ^99mTc-HYNIC-TZ complexes on coligand used.

coligand	LogP
tricine	−0.54 ± 0.07
EDDA	−0.40 ± 0.01
tricine:EDDA	−0.51 ± 0.02

Biology

In vitro studies

To evaluate the bioorthogonal reaction between TCO and tri-Tc-HYNIC-TZ we used the CC49 mAb, which binds to the pancarcinoma antigen TAG72. The antibody TCO conjugation was achieved after a 30 min reaction with a 10 equivalent excess of succinimidyl-TCO (TCO-NHS) as described elsewhere [19]. The number of reactive TCO present per antibody was determined by SDS-PAGE after a 30 min incubation of the conjugated antibody with different molar ratios of tri-^99mTc-HYNIC-TZ resulting in an average of 8.5 TCO per antibody (see Supplementary material, Figure S3).

After TCO conjugation, the immunoreactivity of the modified antibody was determined using immobilized mucin from bovine submaxillary glands (BSM) as the target antigen on high binding ELISA plates with bovine serum albumin (BSA) as a negative control. No significant differences were observed in the antigen recognition between the non-modified and the TCO modified CC49 (Figure 3A).

Figure 3.

Characterization of CC49-TCO’s bioactivity and cycloaddition reactivity. A) Bioactivity assay of CC49-TCO and unmodified CC49 antibody with surface immobilized BSM detected with goat-anti mouse-HRP and ABTS as substrate followed at 405 nm (data represent the mean ± SD. (*) no significant difference (p > 0.05), Student’s t-test). B) Tri-^99mTc-HYNIC-TZ-TCO cycloaddition with BSM surface immobilized CC49-TCO (data represent the mean ± SD. (**) significant difference (p < 0.05), Student’s t-test). C) The reaction of tri-^99mTc-HYNIC-TZ with pretargeted CC49-TCO and CC49 bound to live LS174T cells (data represent the mean ± SD. (**) significant difference (p < 0.05), Student’s t-test).

After confirming that antibody recognition was not affected by TCO conjugation, we evaluated the cycloaddition reaction between tri-^99mTc-HYNIC-TZ and CC49-TCO with a pretargeted approach using immobilized BSM on high binding ELISA plates. The assay consisted of allowing the mAb-TCO conjugate or mAb to bind immobilized BSM followed by the addition of tri-^99mTc-HYNIC-TZ. The tetrazine-TCO reaction was evaluated after 30 min of incubation. The results confirmed that the pretargeted mAb-TCO conjugate was radiolabeled via the tetrazine-TCO reaction and not due to non-specific binding of tri-^99mTc-HYNIC-TZ (Figure 3B).

The next step was to study the application of pretargeting to the detection of the TAG72 biomarker using live cells through the bioorthogonal reaction with tri-^99mTc-HYNIC-TZ. Radioimmunodetection of living LS147T cells was confirmed due to the reaction between bound mAb-TCO and tri-^99mTc-HYNIC-TZ (Figure 3C).

In vivo studies

Biodistribution studies of tri-^99mTc-HYNIC-TZ (6.5×10⁴ Ci/mol) in CD-1 normal mice were performed at 1, 4 and 24 h post-injection (Table 3). Clearance from the blood and major organs was fast except for liver, intestines and kidney. At 1 h post injection of tri-^99mTc-HYNIC-TZ 1.70 ± 0.11 % ID/g was present in blood and 43.76 ± 1.80 % ID was eliminated through bladder and urine. Non-specific accumulation was observed in liver and intestines at 1 h post injection with uptake values of 14.15 ± 0.43 % ID/g and 5.62 ± 0.59 % ID/g, respectively. A significant decrease of activity in the small intestine after 4 h with a concurrent increase to 21.82 ± 2.73 % ID/g in the large intestine suggested that a significant portion of the excretion pathway of the compound was via the gastrointestinal tract, which was probably due to hydrophobic nature of the tri-^99mTc-HYNIC-TZ construct. Urine metabolites were evaluated by HPLC at 1 h post-injection. The HPLC profile showed the presence of a peak in urine with a similar retention time to the tri-^99mTc-HYNIC-TZ (see Supplementary material, Figure S5). Plasma protein binding studies revealed that the percentage of plasma bound protein was 14.0 ± 3.2 % for tri-^99mTc-HYNIC-TZ and 17.3 ± 0.6 % for EDDA complex (see Supplementary material). These are moderate plasma values and are consistent with the blood clearance data in vivo.

Table 3.

Biodistribution of tri-^99mTc-HYNIC-TZ at 1, 4 and 24 h in CD-1 normal mice (n=3).

Percent Injected Dose / gram
Organs	1 h	4 h	24 h
Blood	1.70 ± 0.11	1.01 ± 0.35	0.50 ± 0.02
Heart	0.68 ± 0.06	0.39 ± 0.02	0.33 ± 0.04
Lung	1.33 ± 0.13	0.77 ± 0.07	0.64 ± 0.10
Liver	14.15 ± 0.43†	10.61 ± 0.89†	8.95 ± 0.58
Spleen	0.46 ± 0.04	0.27 ± 0.03	0.25 ± 0.03
Stomach	0.32 ± 0.11	1.49 ± 1.69	0.48 ± 0.16
Large intestine	0.42 ± 0.18	21.82 ± 2.73	1.28 ± 0.28
Small intestine	8.07 ± 0.61	3.45 ± 0.24	0.45 ± 0.01
Intestines	5.62 ± 0.59†	10.01 ± 0.67†	0.71 ± 0.11†
kidneys	7.84 ± 1.63†	3.79 ± 0.51†	3.12 ± 0.25
Brain	0.05 ± 0.01	0.03 ± 0.01	0.02 ± 0.01
Muscle	0.43 ± 0.04	0.24 ± 0.025	0.19 ± 0.03
Bone	0.61 ± 0.09	0.31 ± 0.09	0.49 ± 0.07
Skin	1.41 ± 0.10	0.88 ± 0.08	0.82 ± 0.16
Percent Injected Dose
Bladder and Urine	43.76 ± 1.80	46.01 ± 2.28	74.73 ± 2.68

Statistical analyses was performed using a two way ANOVA with Bonferroni’s

^†p < 0.05).

Evaluation of the tumor imaging properties of pretargeted CC49-TCO and tri-^99mTc-HYNIC-TZ in vivo was performed in mice bearing LS174T colon cancer tumors. For this purpose, CC49-TCO (100 µg) was injected via the tail vein and allowed to accumulate for 48 h. Unbound circulating antibody was reduced using one dose of clearing agent [23], followed 2 h later by the administration of tri-^99mTc-HYNIC-TZ. SPECT/CT imaging was acquired 2 h post injection of radioactivity (Figure 4).

Figure 4.

In vivo SPECT imaging. Coronal slice (A) and transverse slice (C) of pretargeted CC49-TCO in vivo tri-^99mTc-HYNIC-TZ imaging in mice bearing LS174T colon cancer tumor, 2 h post injection. Coronal slice (B) and transverse slice (D) control image of tri-^99mTc-HYNIC-TZ.

The SPECT/CT imaging results showed there was tumor visualization, which confirmed that the in vivo cycloaddition reaction was successful between the pretargeted CC49-TCO and the tri-^99mTc-HYNIC-TZ. Additionally, non-specific background was observed in the whole body image mainly in the liver, gall bladder, intestines and kidneys as predicted by the biodistribution studies and the tri-^99mTc-HYNIC-TZ control image (Figure 4B and D). These results suggested that the hydrophobic nature of tri-^99mTc-HYNIC-TZ resulted in non-specific background and highlighted the need for a more hydrophilic second generation of HYNIC-TZ construct.

A biodistribution study was performed with tri-^99mTc-HYNIC-TZ in LS174T tumor bearing mice at the 2 h imaging time point. A modest amount of tumor uptake (1.39 ± 0.43 % ID/g; Supplemental Table 3) was evident, however the majority of the radioactivity was in the liver and gastrointestinal tract and the kidneys. The results of the tri-^99mTc-HYNIC-TZ biodistribution study were consistent with the imaging study. A blocking experiment was performed in which a 10-fold excess of the CC49 mAb was co-injected with the CC49-TCO mAb. A 17 % reduction in tumor uptake was observed (data not shown). Tumor uptake of ¹²⁵I iodinated CC49-TCO was high at 43.53 % ID/g (Supplemental Table 3). The low tumor uptake of the tri-^99mTc-HYNIC-TZ was not due to a lack of CC49-TCO mAb but likely the result of being sequestered in the liver and GI tract and unavailable for reaction with tumor targeted CC49-TCO mAb.

Pretargeted SPECT and PET in vivo imaging using the Diels-Alder TCO-Tz cycloaddition approach have been successful in preclinical studies with ¹¹¹In [19], ¹⁷⁷Lu [23], ⁶⁴Cu [49] and ¹⁸F [50]. All of these examples have involved tetrazines linked to a cyclic aminopolycarboxilic acid metal chelators like DOTA or NOTA, which appears to provide sufficient solubility for favorable in vivo imaging. The ^99mTc-HYNIC and ^99mTc-tricarbonyl complexes are far less hydrophilic than the radiometal cyclic aminopolycarboxilic acid complexes. Improving the hydrophilicity of the tri-^99mTc-HYNIC-TZ is necessary to achieve desirable in vivo biodistribution. The addition of a polyethylene glycol spacer and or a charged amino acid peptide sequence between the HYNIC and TZ should increase the hydrophilicity of the HYNIC-TZ conjugate and improve its in vivo pharmacokinetic properties resulting in high tumor to background ratios.

Conclusions

We have developed a HYNIC-tetrazine derivative with the hydrazino moiety protected by a trifluoroacetyl moiety. The construct was labeled with ^99mTc efficiently under mild conditions. The radiolabeled compound was shown to be stable in vitro and in vivo. The tri-^99mTc-HYNIC-TZ and TCO interaction was evaluated, showing promising results in vitro. In vivo, the pretargeted ^99mTc-based SPECT tumor imaging proof of concept was positive. However, the biodistribution profile and imaging studies indicated the need of a more hydrophilic tri-^99mTc-HYNIC-TZ derivative.

Experimental

General

All chemicals and solvents were purchased from Sigma-Aldrich, Merck, Dorwill and Carlo Erba. All solvents for organic synthesis were distilled prior to use. Analytical TLC was performed on alumina plates and visualized with UV light (254 nm). (E)-4-Cycloocten-1-yl 2,5-dioxo-1-pyrrolidinyl ester carbonic acid (TCO-NHS) was purchased from Click Chemistry Tools, LLC., and 4-(6-methyl-1,2,4,5-tetrazin-3-yl)phenylmethanamine hydrochloride was purchased from Conju-Probe, LLC, San Diego. Structural elucidations were based on ¹H, ¹³C, COSY, HMBC, and HSQC spectroscopies, and MS spectrometry. NMR spectra were acquired on a Bruker DPX-400 spectrometer. The chemical shifts values were expressed in ppm relative to tetramethylsilane as internal standard. Mass spectra were determined on a Thermo LCQ Fleet LC/MS spectrometer using electrospray ionization (ESI). To determine the purity of the compound, elemental microanalyses obtained on a Carlo Erba Model EA1108 elemental analyzer from vacuum-dried sample was used. The analytical results for C, H, and N were within ± 0.4 of the theoretical values. Water was purified and deionized (18 MΩ/cm²) on a Milli-Q water filtration system (Millipore Corp., Milford, MA). ⁹⁹Mo-^99mTc generators were purchased from TecnoNuclear (Argentina). Radioactivity was counted in a CRC7 Capintec dose calibrator and in a solid scintillation counter detector with 3"×3" NaI(Tl) crystal associated with a single channel analyzer (ORTEC, Oak Ridge, TN). High performance liquid chromatography (HPLC) was performed on an Agilent 1200 Series Infinity Star equipped with GABI detector, a UV detector and a ThermoScientific Hypersil ODS reverse phase C18 column (300 mm x 4.6 × 10 microns). Radio-TLC detection was accomplished using instant thin-layer chromatography (ITLC) on silica gel strips (Pall Corporation, Port Washington, NY).

Organic Syntheses

N-Succinimidyl 6-(trifluoroacetyl)hydrazinopyridine-3-carboxylic acid

The product was synthesized as described elsewhere [48].

N’-{5-[4-(6-methyl-1,2,4,5-tetrazin-3-yl)phenylmethylcarbamoyl]pyridin-2-yl}-2,2,2-trifluoroacetohydrazide (1)

To a mixture of 4-(6-methyl-1,2,4,5-tetrazin-3-yl)phenylmethanamine hydrochloride (20 mg, 84.14 µmol) and N,N-diisopropylethylamine (2 equivalents) in a mixture of THF:DMF (4:1) (4 mL), 1.2 equivalents of N-succinimidyl 6-(trifluoroacetyl)hydrazinopyridine-3-carboxylic acid was added. The reaction mixture was stirred at room temperature for 24 h. The solvent was concentrated in vacuo and the final product purified on a preparative alumina TLC using CH₂Cl₂:MeOH (95:5) as mobile phase. After evaporation in vacuo (1) was obtained as pink solid (10 mg, 27 % yield, mp: 242–243 °C). The purity of (1) was verified by HPLC (Figure S1). ¹H NMR (400 MHz, DMSO-d₆) δ (ppm): 9.40 (bs, 2H, NHNH), 8.81 (bt, J = 4.8 Hz, 1H, NHCOPy), 8.49 (s, 1H, Py), 8.43 (d, J = 8.2 Hz, 2H, Ar), 7.91 (d, J = 8.8 Hz, 1H, Py), 7.58 (d, J = 7.9 Hz, 2H, Ar), 6.75 (d, J = 8.4 Hz, 1H, Py), 4.55 (d, J = 4.8 Hz, 2H, CH₂), 2.98 (s, 3H, CH₃). ¹³C NMR (101 MHz, DMSO-d₆) δ (ppm): 167.22 (CCH₃), 165.86 (Py-CO), 163.15 (Ar-CN₂), 158.20 (q, J = 31 Hz, C=OCF₃), 158.08 (Py), 148.56 (Py), 145.13 (Ar), 136.48 (Py), 130.40 (C₂N₄-C(Ar)), 128.23 (Ar), 127.59 (Ar), 117.18 (Py), 113.01 (q, J = 297 Hz, C=OCF₃), 104.36 (Py), 42.36 (CH₂), 20.97 (CH₃). ESI-MS, expected m/z: 433.13 (M+H)^+., found m/z: 433.12 (M+H)^+.. Anal. calc for C₁₈H₁₅F₃N₈O₂: C, 50.0; H, 3.5; N, 25.9. Found: C, 49.9; H, 3.2; N, 25.6.

^99mTc-HYNIC-tetrazine radiolabeling

Tricine as coligand (100 µL of tricine solution in water (100 mg/mL), 55.8 µmol) was added to 20 µg of (1) (1 µg/µL, 0.46 nmol), followed by addition of 20 µL of a fresh solution of SnCl₂ (2 mg/mL in ethanol, 17.7 µmol) and 0.111 MBq of Na^99mTcO₄ (up to 200 µL). The mixture (pH 5.0) was incubated for 15 min at 37 °C in a dry bath. Labeling efficiency was determined using HPLC with (A) water/0.05% TFA and (B) MeOH as mobile phases (0% to 100% B in 20 min and 100% to 0% B in 2 min) and ITLC-SG (citrate 0.1M) to quantify ^99mTc-colloidal.

Similar experimental conditions were employed for the preparation of ^99mTc-HYNIC-tetrazine using ethylenediaminediacetic acid (EDDA) or a mixture of tricine:EDDA as coligands, respectively. In the case of EDDA alone as coligand, 7.1 µmol was used in the procedure. For the mixture 55.8 µmol of tricine and 7.1 µmol of EDDA was used. See supplementary information for details of experimental conditions.

Physicochemical studies

Partition coefficient

Log P values were determined in triplicate as follows. The radiolabeled products were purified by RP-HPLC with C18 column and water (0.1 % TFA) and MeCN (0.1 % TFA) as the mobile phases (0% to 3% B in 3 min, 3% to 25% B in 0.1 min, 25% to 30% in 20 min, 30% to 90% in 2 min, 90% 3 min, 90% to 3% in 2 min, 3% 5 min) followed by C18 SEP-PAK cartridge purification with 50 % ethanol in phosphate buffered saline pH 7.4 (PBS) and purged with N₂. An appropriate amount of the corresponding ^99mTc-HYNIC-tetrazine (6475 MBq/mg) was dissolved in PBS to get a final concentration of 200,000 CPM/mL. A mixture of 500 µL of n-octanol and 500 µL of ^99mTc-HYNIC-tetrazine diluted in PBS was mixed vigorously for 1 min and centrifuged at 14,000 rpm for 10 minutes. Three fractions of 100 µL were collected from both phases of each tube, and the radioactivity counted in a NaI well counter. The partition coefficient was obtained as log₁₀(n-octanol counts/aqueous phase counts).

In vitro stability of ^99mTc-HYNIC-tetrazine radiolabeling

In vitro stability of tri-^99mTc-HYNIC-TZ was evaluated in the reaction mixture and in PBS at 25 °C during 2, 4 and 24 h post labeling in triplicate. The labeling yield was determined using RP-HPLC and ITLC-SG (citrate 0.1 M) to quantify ^99mTc-colloidal.

In vitro biological studies

Antibody production

The CC49 mAb was purified from CC49 hybridoma (ATCC) culture media using protein-A affinity chromatography (MABTRAP, GE Health Sciences). The purified antibody was concentrated to 5 mg/mL in PBS, aliquoted and stored at −80° C prior to use.

Antibody TCO conjugation

To 2 mg of CC49 antibody (5mg/mL solution) in PBS, 15 µL of sodium carbonate buffer (1 M) was added to adjust the pH of the mixture to 9.5, followed by the addition of TCO-NHS (17 µg in 3.4 µL of dry DMSO, 10 molar excess). The reaction mixture was incubated at room temperature for 30 min with shaking in darkness. The TCO conjugated mAb was purified from free NHS-TCO reagent using a PD-10 column (Amersham Biosciences) with PBS. The concentration of purified mAb-TCO was determined by measuring UV absorbance at 280 nm.

CC49-TCO receptor recognition

BSM (10 µg/mL ) in PBS was immobilized on 96 well ELISA plate overnight at 4 °C and blocked with blocking buffer (Rockland) for 1 h at 37°C. BSA (10 µg/mL) was used as a negative control. After several washes with 0.1 % PBS-Tween, 100 µL of CC49 and CC49-TCO (10 µg) were incubated in BSM and BSA coated wells for 1 h at 37 °C and washed with 0.1 % PBS-Tween (three times). The binding of CC49 antibody to immobilized antigen was detected colorimetrically using an HRP-conjugated anti-mouse antibody and 2,2’-azino-bis(3-ethylbenzothiazoline-6-sulphonic acid) (ABTS) as the substrate.

Tetrazine-TCO reaction of surface of immobilized CC49-TCO

BSM (10 µg/mL) was immobilized on 96 well ELISA plate overnight at 4 °C and blocked with blocking buffer (Rockland) for 1 h at 37°C. BSA (10 µg/mL) was used as a control. After several washes with (PBS - 0.1 % Tween), 100 µL of CC49 and CC49-TCO (10 µg) were incubated in BSM and BSA coated wells for 1 h at 37 °C, followed by 3 washes with 0.1 % PBS-Tween. Subsequently, 100 µL of tri-^99mTc-HYNIC-TZ (1µCi /100 µL) in BSA (1 %) was incubated for 30 min and washed three times with 0.1 % PBS-Tween. Finally the radiolabeled antibody-BSM complex was removed with 300 µL of aqueous NaOH (2 M), collected and measured on a gamma counter.

Pretargeted LS147T cell binding

LS147T cells in RPMI media with fetal bovine serum (10 %) were seeded (0.5 × 10⁶) in 24 well plates and allowed to grown overnight. Live cells were washed with PBS (1 % BSA) and incubated for 30 min with 200 µL of CC49-TCO and CC49 (100 µg /mL). After the incubation period, the cells were washed with 500 µL of PBS and incubated for 30 min with 1 µCi of tri-^99mTc-HYNIC-TZ (5.5 MBq/µg, in 200 µL of PBS. After two washes with PBS the cells were lysed with aqueous NaOH (2 M). The milieu was collected and radioactivity measured on a gamma counter to quantify membrane bound activity.

In vivo biological studies

Animal studies were conducted in compliance with the University of Missouri Institutional Animal Care and Use Committee approval.

Tumor model

The tumor cell line LS147T was purchased from ATCC and grown in RPMI media with fetal bovine serum (10 %). Nude (nu/nu: Harlen) mice were inoculated subcutaneously with 2 million LS 147T cells (PBS:Matrigel, 2:1). Biodistribution and imaging studies were performed when the tumors reached 0.5–1 cm³.

In vivo biodistribution of ^99mTc-HYNIC-tetrazine

Tumor free CD-1 mice were injected intravenously through the tail vein with tri-^99mTc-HYNIC-TZ (0.17–0.74 MBq, 5.55 GBq/mg) and sacrificed after 1, 4 and 24 h post injection (n=3). The organs and tissues of interest were dissected, weighed and the radioactivity of the samples was measured in a gamma counter. Radioactivity in the urine and feces was also determined. The radioactivity was expressed as percentage of injected dose per gram of tissue (% ID/g) or percent of injected dose (% ID). The total amount of blood was considered as 6.5 % of total body weight.

Pretargeted in vivo imaging

Tumor bearing mice (n = 2) were injected via the tail vein with 100 µg of CC49-TCO. Forty-eight hours post antibody administration, the mice were injected with 160 µg of clearing agent (Tetrazine and galactose conjugated albumin, Tagworks Pharmaceuticals, The Netherlands [23]) followed by the intravenously administration of tri-^99mTc-HYNIC-TZ (1:1 molar ratio with TCO:(1), 7.4 MBq/µg ~21 MBq in 100 µL). A control group of mice received tri-^99mTc-HYNIC-TZ alone. After 2 h the mice were anesthetized with 1 % – 2 % isofluorane and SPECT-CT images were acquired with a Siemens Inveon SPECT/CT unit (Siemens Medical Solution Incorporated, TN, USA) equipped with dual pixellated NaI detectors, 1.0 mm multi-pinhole collimators, a CCD X-ray detector, and an 80 kVp microfocus X-ray source. Volumetric SPECT data were generated with a three-dimensional ordered-subsets expectation maximization (OSEM) algorithm with geometric misalignment corrections. Concurrent microCT whole-body imaging was performed to allow for fusion of anatomic and molecular data.

Abbreviations

TCO

trans-cyclooctene

TZ

1,2,4,5-tetrazine

PET

Positron Emission Tomography

HYNIC

6-hydrazinonicotinc acid

SPECT

Single Photon Emission Computed Tomography

PBS

phosphate buffer saline

BSA

bovine serum albumin

ABTS

2,2’-azino-bis(3-ethylbenzothiazoline-6-sulphonic acid)

RCP

radiochemical purity

EDDA

ethylenediaminediacetic acid

DMSO

dimethyl sulfoxide

SDS-PAGE

sodium dodecyl sulfate polyacrylamide gel electrophoresis

BSM

bovine submaxillary glands

CPM

counts per minute

mAb

monoclonal antibody