Development of a 64Cu-labeled CD4+ T cell targeting PET tracer: evaluation of CD4 specificity and its potential use in collagen-induced arthritis

Mice

Female DBA/1JRj mice (7 weeks) and C57BL/6JRj (7 weeks) were purchased (Janvier, Le Genest-Saint-Isle, France) and housed in groups of 4–8 mice in individually ventilated cages under standardized lighting conditions. Mice were fed pathogen-free food and water ad libitum. All animal experiments were conducted under protocols approved by the National Animal Experiments Inspectorate under the license number 2016-15-0201-00920.

Collagen-induced arthritis (CIA)

Mice were immunized subcutaneously at the base of the tail (100 µL) with chicken CII (Chondrex, Redmond, WA, USA) emulsified in complete Freund's adjuvant (Sigma-Aldrich, St. Louis, MO, USA). Three weeks after initial immunization, mice received a booster immunization with CII emulsified in incomplete Freund's adjuvant (Sigma-Aldrich, St. Louis, MO, USA) (100 µL) [11]. CII was dissolved O/N at 4 °C in 10 mM acetic acid. Mice were anesthetized in 3–4% sevoflurane in 65% N2 and 35% O2 during both injections. Signs of arthritis in the paws appeared approximately four weeks after initial immunization. Each paw was visually scored three times a week on a scale from 0 to 4. The following criteria were used: 0: normal paw, 1: one toe inflamed and swollen, 2: more than one toe, but not entire paw, inflamed and swollen or mild swelling of entire paw, 3: entire paw inflamed and swollen and 4: very inflamed and swollen paw or ankylosed paw [12]. For ethical purposes, a mean score of 2.5 for all paws was considered maximum per animal. Mice were euthanized if exceeding the mean score. Data are presented as the mean score of all four paws. In case of lameness, score ≥ 2, the mice were treated with daily analgesia (buprenorphine, 1 mg/kg, intraperitoneal (i.p.)). For therapeutic assessment, the CIA mice were treated daily with DXM for 7 days starting on day 28 after initial immunization [13,14,15]. DXM was administered as i.p. injections in a total volume of 200 µL/mouse (1 mg/kg in NaCl).

Radiosynthesis of [64Cu]Cu-NOTA-CD4 and isotype control

F(ab)’2 fragments were generated from the GK1.5 monoclonal antibody R-anti-mouse CD4 (#BP0003-1, Bio X cell, Lebanon, NH, USA) and R-anti-mouse IgG2b clone LTF-2 (#BE0090, Bio X cell) using FabRICATOR enzyme (#A0-FR1-050, Genovis, Lund, Sweden). 2.5 mg CD4-F(ab)′2 and IgG2b-F(ab)’2 in PBS was incubated with 500 units FabRICATOR for 2.5 h at 37 °C. The crude antibody–enzyme mixture was purified by preparative HPLC (Yarra-2000 SEC column, 0.1 M phosphate buffer, 1 ml/min) yielding isolated F(ab)’2 and Fc fragments. The fragments purified followed by conjugation to the 2-S-(isothiocyanatbenzyl)-1,4,7-triazacyclononane-1,4,7-triacetic acid (p-SCN-Bn-NOTA) chelator, as previously described [16].

[64Cu]Cl2 (DTU Nutech, DTU) was dissolved in TraceSelect water (Merck Millipore) to a final concentration of 2 GBq/mL. NOTA-CD4-F(ab)’2 (100 µg in PBS) was incubated with 250 MBq [64Cu]Cl2 in 0.1 M NaoAc buffer pH = 5.5 with 5 mg/ml gentisic acid (15 min, 37 °C). The reaction was quenched with 10 mM EDTA followed by PD10 purification into PBS. The radiochemical yield and purity at end-of-synthesis (EOS) were determined by size-exclusion-chromatography-HPLC (SEC-HPLC) using an isocratic method with 0.1 M phosphate buffer pH = 7 as mobile phase and a flow rate of 1 ml/min.

SDS-PAGE

Full-length monoclonal CD4 antibody (IgG), reduced full-length CD4 antibody and the tracer [64Cu]Cu-NOTA-CD4 were mixed with NuPAGE LDS sample buffer (#NP0007, Invitrogen, Carlsbad, CA, USA) and denatured for 10 min at 70 °C and loaded onto BoltTM 4–12% Bis–Tris gels (#NW04120, Invitrogen). Electrophoresis was run on the Mini Gel Tank (Thermo Fisher Scientific, Waltham, MA, USA) at 200 V constant voltage in NuPAGE MES SDS running buffer (#NP0002, Invitrogen). The SDS-PAGE gel was fixed and stained using Coomassie brilliant blue R-250 (#1610436, Bio-Rad, Hercules, CA, USA). The gel was analyzed for radioactive content by exposure to Multisensitive Phosphor Screens and imaged using the Amersham Typhoon Imaging System (GE Healthcare, Chicago, IL, USA).

Mouse T cell isolation, immunoreactivity and saturation binding assay

Murine CD4+ cells were isolated from mouse spleen using the magnetic activating cell sorting (MACS) technique (CD4+ T cell isolation kit, #130-104-454; CD8a+ T cell isolation kit, #130-104-075, Miltenyi Biotec). Female DBA/1JRj mice (7 weeks old) were killed by cervical dislocation and the spleens removed aseptically.

The immunoreactivity of anti-CD4-F(ab)’2 fragments following radiolabeling was assessed according to the Lindmo assay [17]. Increasing concentrations of CD4+cells (2.5 × 106–4 × 107 cells/mL) were incubated with 1 nM [64Cu]Cu-NOTA-CD4 for 3 h at 4 °C. Cells were centrifuged at 500 g for 5 min and the supernatants and pellets counted in a gamma counter (Wizard2, PerkinElmer, Massachusetts, USA). Cell-associated radioactivity was calculated as the ratio of cell-bound radioactivity to the total amount of added radioactivity.

The affinity of radiolabeled anti-CD4-F(ab)’2 was assessed by a saturation binding assay. T cells were harvested as described above, added in triplicates (2 × 104 cells) to a MultiScreenHTS BV Filter Plate 1.2 µm (#MSBVN1250, Merck Millipore) and washed twice in PBS. Eight different concentrations of [64Cu]Cu-NOTA-CD4 [range: 60–0.04 nM] in PBS supplemented with 1% bovine serum albumin (BSA) were added the wells. The plate was incubated for 4 h at 4 °C. After incubation, the plate was washed 3 times in PBS with 1% BSA using a vacuum manifold (Macherey-Nagel, Fisher Scientific). The plastic cover was removed from the plate bottom and the plate dried in a heat cabinet. The dry filters were transferred to counting tubes and counted in a gamma counter.

In vivo PET/CT imaging

Mice were intravenous (i.v.) injected with [64Cu]Cu-NOTA-CD4 and 30 µg CD4-F(ab)’2 co-dose in a total volume of 150 µL (diluted in PBS) in the tail vein as a single bolus. Co-doses was given to decrease binding to lymphoid tissue significantly, but without blocking the specific uptake. The co-dose was based on previously reported data where the optimal dose was found to be 30 µg [16]. Whole-body imaging consisted of a CT acquisition followed by a static PET acquisition and was performed on the Inveon® small animal imaging system (Siemens Medical Systems, Malvern, PA, USA). Mice were anesthetized in 3–4% sevoflurane in 65% N2 and 35% O2 prior to imaging. All CIA mice were injected with tracer on day 35 after initial immunization. For longitudinal imaging evaluation, the PET/CT imaging was performed 1, 4, 24 and 44 h post-injection (p.i.) of [64Cu]Cu-NOTA-CD4 (2.52 ± 0.04 MBq, 0.5 ± 0.02 µg) and 30 µg CD4-F(ab)’2 co-dose in CIA (n = 6) and control mice (n = 4). For evaluation of DXM, PET/CT imaging was performed 24 h p.i. on day 36 after initial immunization of CIA (n = 23) and DXM-treated mice (n = 12) (four independent experiments: EXP1-4). Untreated control mice (n = 11) were also included. All mice received [64Cu]Cu-NOTA-CD4 (EXP1: 1.26 ± 0.02 MBq, 0.4 ± 0.02 µg; EXP2: 7.51 ± 0.29 MBq, 0.55 ± 0.03 µg; EXP3: 1.86 ± 0.10 MBq, 0.38 ± 0.01 µg; EXP4: 2.33 ± 0.02 MBq, 0.41 ± 0.01 µg) and 30 µg CD4-F(ab)’2 co-dose. For isotype control imaging, CIA (n = 8) and control mice (n = 6) were i.v. injected with [64Cu]Cu-NOTA-IgG2b, an isotype-matched non-specific control IgG antibody fragment labeled with 64Cu (1.95 ± 0.04 MBq, 0.39 ± 0.01 µg) and IgG2b-F(ab)’2 30 µg co-dose in a total volume of 150 µL (diluted PBS) in the tail vein as a single bolus. The mice were subjected to PET/CT imaging with 900 s static PET acquisition 24 h p.i. according to the previous described protocol. Whole-body irradiation (WBI) (n = 8/group) and non-irradiated control mice (n = 6/group) were either injected with [64Cu]Cu-NOTA-CD4 (2.40 ± 0.03 MBq, 0.60 ± 0.01 µg) or [64Cu]Cu-NOTA-IgG2b (2.25 ± 0.04 MBq, 0.56 ± 0.01 µg) and 30 µg co-dose. PET data were acquired in list mode with an acquisition time of 300, 300, 600 and 1200 s for the 1, 4, 24 and 44 h time points, respectively. PET scans were reconstructed using a maximum a posteriori algorithm with CT-based attenuation correction.

Imaging analysis

Image analysis was performed using the Inveon® Research Workstation software (Siemens Medical Systems, PA, USA). A CT-based region of interest (ROI) tool was used to carefully draw around each carpal and tarsal joint, forming four ROIs per mouse. ROIs were also drawn over the heart, kidney, spleen, thymus, liver and thigh muscle. The activity in the blood was calculated as 20% maximum activity in the heart. The uptake of [64Cu]Cu-NOTA-CD4 and [64Cu]Cu-NOTA-IgG2b was quantified as percent injected dose per joint (%ID/joint) and maximum percent injected dose per gram tissue (%Max ID/g) assuming a soft tissue density of 1 g/cm3. Target-to-blood ratios of [64Cu]Cu-NOTA-CD4 uptake were calculated as maximum uptake (%Max ID/g) in the most affected joints (score 3) divided by mean uptake in the blood (%ID/g) to determine optimal scanning time. Moreover, joint-to-blood ratios were calculated for the DXM assessment.

Whole-body irradiation (WBI)

WBI was performed to test the effect of irradiation on CD4+ T cells in spleen for evaluation of tracer specificity. Mice were exposed to a single γ-radiation dose of 5 Gy, 1 Gy/min (320 kV, 12.5 mA) using an X-RAD 320 (PXi, North Branford, CT, USA). Two groups of C57BL/6JRj mice (20 weeks) received WBI the day before injection of [64Cu]Cu-NOTA-CD4 (n = 8) or [64Cu]Cu-NOTA-IgG2b (n = 8). Two groups of healthy non-irradiated C57BL/6 control mice (n = 6/group) were also included.

Immunohistochemical analysis

CIA and control mice were euthanized by cervical dislocation and the carpal and tarsal joints were isolated. Joints were decalcified for three weeks in 10% formic acid in buffered formaldehyde 4% followed by preparation in Shandon Excelsior AS Tissue Processor (Thermo Fisher Scientific, Waltham, MA, USA) O/N and embedded in paraffin. Paraffin-embedded joints were sectioned at 4 µM and dewaxed through xylene to tap water. For antigen retrieval, sections were boiled in microwave for 15 min in 10 mmol citrate buffer (pH 6) and pre-incubated in 2% bovine serum albumin (BSA) for 10 min followed by incubation with primary recombinant anti-CD4 antibody (#ab183685, Abcam, Cambridge, UK) at 1:500 dilution in 2% BSA O/N at 4 °C. Sections were incubated for 40 min with biotinylated secondary goat-anti-rabbit IgG antibody (BA-1000, Vector Laboratories, Burlingame, CA, USA) at 1:200 dilution. Afterward, 3% hydrogen peroxide blocked the endogenous peroxidase. To amplify the reaction, sections were incubated with Avidin and Biotinylated horseradish peroxidase macromolecular Complex (ABC-Elite) (PK-6100, Vector Laboratories) for 30 min. Finally, the reaction was developed by the use of 3,3–diaminobenzidine (SK-4100, Vector Laboratories) for 15 min and counterstaining was performed with Mayer’s Hematoxylin Solution (Sigma-Aldrich, St. Louis, MO, USA). All procedures were performed at RT if not stated otherwise. Sections were stained in the same analysis. Images were taken using an Olympus BX51 microscope with a XC-10 camera.

Flow cytometry

Joint-infiltrating cells were isolated by a procedure adapted from [18]. Briefly, the carpal and tarsal joints were isolated. Skin and fur were removed, and joints were mechanically dissociated using surgical scissors followed by digestion in RPMI 1640 supplemented with type 1-S hyaluronidase from bovine testes (2.4 mg/mL) (Sigma, St. Louis, MO, USA), collagenase VIII (1 mg/mL) (Sigma), 10 mM HEPES, 50 units/mL penicillin and 50 µg/mL streptomycin for 1 h at 37 °C in a shaking water bath. Digested joints were passed through a 70 µm cell strainer.

Processed cells were blocked with Fc-block (clone 2.4G2) (BD Biosciences, San Jose, CA, USA) in FACS buffer (PBS + 0.5% BSA + 0.1% NaN3 + 2 mM EDTA) for 5 min on ice. Fc-blocked samples were stained for 30 min on ice in a master mix of FACS buffer, brilliant stain buffer, amine reactive dye (eFluor780 viability dye) (Thermo Fisher Scientific, Waltham, MA, USA) and the following antibodies: anti-CD3 (clone 145-2C11) PE-CF594, anti-CD4 (clone RM4-5) BV786, anti-CD8 (clone 53–6.7) APC, anti-CD11b (clone M1/70) BV480, anti-CD11c (clone N418) PE-Cy7, anti-CD45 (clone 30-F11) AF700, anti-CD64 (clone X54-5/7.1) BV421, anti-I-A/I-E (clone M5/114.15.2) and anti-Ly6g (clone 1A8) BV711. Antibodies against CD3, CD4, CD8, CD11b, CD45, CD64 and Ly6g were purchased from BD Biosciences. Antibodies against CD11c and I-A/I-E were purchased from BioLegend (BioLegend, San Diego, CA, USA). Cells were acquired on a BD LSRFortessa X-20 using BD FACSDiva v.8.0.1 software (BD Biosciences, San Jose, CA, USA). Data analysis was performed in FlowJo v.10.6.1, Tree Star Inc. (FlowJo, Ashland, OR, USA).

Statistical analysis

One-way analysis of variance with Tukey’s post hoc test was used to assess statistically significant differences between groups. p values < 0.05 were considered statistically significant. Prism 8.0c (GraphPad Software, La Jolla, CA, USA) was used for all statistical analysis.

留言 (0)

沒有登入
gif