Published online before print January 9, 2008, 10.1148/radiol.2463070471
(Radiology 2008;246:854-862.)
© RSNA, 2008
Detection of Cell Death in Tumors by Using MR Imaging and a Gadolinium-based Targeted Contrast Agent1
Anant S. Krishnan, BSc, MBBS, MRCS,
Andre A. Neves, MEng, PhD,
Maaike M. de Backer, PhD,
De-En Hu, MD,
Bazbek Davletov, PhD,
Mikko I. Kettunen, PhD, and
Kevin M. Brindle, BA, DPhil
1 From the Department of Biochemistry, University of Cambridge, Tennis Court Road, Cambridge CB2 1GA, England (A.S.K., A.A.N., M.M.d.B., D.E.H., M.I.K., K.M.B.); and Neurobiology Division, MRC Laboratory of Molecular Biology, Cambridge, England (B.D.). Received March 12, 2007; revision requested May 17; revision received June 11; accepted July 18; final version accepted September 12. Supported by a grant from Cancer Research UK (CUK grant C197/A3514) and the Cambridge-MIT Institute. A.S.K. supported by a Medical Research Council (UK) studentship.
Address correspondence to K.M.B. (e-mail: kmb{at}mole.bio.cam.ac.uk).
 |
ABSTRACT
|
|---|
Purpose: To prospectively determine in an animal model whether an ionic gadolinium (Gd3+) chelate conjugate of the C2A domain of synaptotagmin I can be used with magnetic resonance (MR) imaging to detect tumor cell death noninvasively in vivo.
Materials and Methods: Animal experiments were approved by a local ethics review committee. Gd3+ chelates and fluorescent probes were attached to the lysine 
-amino groups of a glutathione-S-transferase–C2A fusion protein. Binding to phosphatidylserine (PS) was characterized by using surface plasmon resonance, and binding to dying cells in vitro was characterized by using flow cytometry and MR imaging. Binding to dying tumor cells in vivo was detected with T1 mapping and T1-weighted MR imaging and compared in drug-treated animals (n = 10); in animals injected with a site-directed mutant, which was inactive in PS binding (PS inactive) and which showed lesser binding to dying cells (n = 6); and in untreated animals injected with PS-active (n = 6) and PS-inactive (n = 6) contrast agents. Among groups, differences that were significant were analyzed by using analysis of variance and Dunnett post hoc analysis.
Results: The contrast agent had a relatively high affinity for PS (dissociation constant = 333 nmol/L ± 85 [mean ± standard error of the mean]; n = 3) and bound to apoptotic and necrotic, but not viable, cells in vitro. There was a greater tumor accumulation of the PS-active contrast agent compared with the PS-inactive contrast agent in drug-treated animals (P < .05) and compared with untreated animals injected with the PS-active and PS-inactive contrast agents (P < .01 for both).
Conclusion: A relatively small (approximately 100 kDa) Gd3+-based contrast agent, which gives positive contrast on MR images, can be used to detect tumor cell death in vivo, and future derivatives of it may be used to assess early tumor responses to treatment.
Supplemental material: http://radiology.rsnajnls.org/cgi/content/full/2463070471/DC1
© RSNA, 2008
|
INTRODUCTION
|
|---|
Tumor responses to treatment have conventionally been assessed from imaging measurements of tumor volume (1); however, tumor shrinkage may take several weeks to manifest. The introduction of new drugs, and subsequent selection of treatment regimens in the clinic, would benefit from techniques that give earlier indications of positive responses to treatment. One such approach is to detect tumor cell death or apoptosis, for the early onset of apoptosis has been shown to be a good prognostic indicator of treatment outcome (2–5).
Apoptotic and nonapoptotic modes of cell death, which can occur in tumors after drug treatment, frequently result in the exposure of the plasma membrane phospholipid phosphatidylserine (PS) (6,7). Apoptotic cells express PS on their surface, whereas in necrotic cells the plasma membrane becomes permeable to the contrast agent, resulting in exposure of PS on the inner leaflet of the plasma membrane bilayer. PS is bound with high affinity by the protein annexin A5 (8), and several annexin A5–based probes have been developed for the noninvasive detection of cell death in vivo by using radionuclide, optical, and magnetic resonance (MR) imaging techniques (9–13). We previously introduced the C2A domain of synaptotagmin I as an alternative PS-targeted contrast agent. At 14 kDa, it is much smaller than annexin A5 (14), and we showed that it could be used to detect cell death in vivo by using MR imaging (15). More recently, it has also been used for radionuclide imaging of cell death (16). For these previous MR imaging measurements, we used a superparamagnetic iron oxide (SPIO) nanoparticle conjugate of the protein. Although sensitive to MR imaging detection, the SPIO label is relatively large (diameter, approximately 30 nm), which can limit access to, and clearance from, the tumor interstitium, thus reducing tissue contrast. The label also gives rise to a decrease in signal intensity (negative contrast), making it more difficult to detect in tumors, which often have preexisting regions of low signal intensity.
Ionic gadolinium (Gd3+) chelate conjugates of C2A can be both smaller than SPIO and give increased signal intensity (positive contrast) on T1-weighted MR images, making them easier to detect. Previously, we showed that biotinylated C2A, attached through the biotin moiety to Gd3+ chelate–conjugated avidin, had sufficient sensitivity to allow the MR detection of apoptotic cells in vitro (17). The purpose of this study was to prospectively determine in an animal model whether a Gd3+ chelate conjugate of the C2A domain of synaptotagmin I can be used with MR imaging to detect tumor cell death noninvasively in vivo.
|
MATERIALS AND METHODS
|
|---|
All animal experiments were conducted in accordance with the Animals (Scientific Procedures) Act of 1986 (United Kingdom) and were designed with reference to the UK Co-ordinating Committee on Cancer Research Guidelines for the Welfare of Animals in Experimental Neoplasia. The work was approved by a local ethics review committee.
Preparation of Contrast Agents
Glutathione-S-transferase (GST)–C2A fusion protein (GST-C2A) and a site-directed mutant of GST-C2A (D230N), which is inactive in PS binding (PS-inactive GST-C2A), were covalently linked to the chelating agent diethylenetriaminepentaacetic acid (DTPA) and fluorescein by means of a reaction of their lysine -amino groups with S-2-(4-isothiocyanatobenzyl)-DTPA, also called p-SCN-Bn-DTPA, and fluorescein isothiocyanate (FITC) (Sigma-Aldrich, Gillingham, Dorset, England), respectively. These procedures were performed by two authors (A.S.K. and A.A.N., with 3 years of experience with use of these reagents). The proteins were expressed in Escherichia coli and purified as described previously (17). Gd3+ conjugates were prepared by using a modification of a protocol described in a study by Niemi et al (18). Both proteins possess 35 lysine residues, 14 on C2A and 21 on GST. The number of Gd–p-SCN-Bn-DTPA molecules conjugated to each protein was estimated to be between 10 and 14. Because GST-C2A exists as a dimer (approximately 85 kDa), these modifications produced conjugates of approximately 90–100 kDa. The Gd3+ conjugates and the unmodified proteins were also labeled, for some experiments, with FITC. The T1 relaxivity values of the PS-active and PS-inactive contrast agents ranged between 50 and 60 (mmol/L)–1 · sec–1 and 50–100 (mmol/L)–1 · sec–1, respectively, at 9.4 T (Appendix E1 [http://radiology.rsnajnls.org/cgi/content/full/2463070471/DC1]).
Surface Plasmon Resonance
Protein affinity for PS-containing liposomes was estimated as described previously (17), and measurements were performed by one author (A.A.N.).
Protein Binding to Large Multilamellar Vesicles
Large multilamellar vesicles (LMVs) containing PS, phosphatidylethanolamine, and phosphatidylcholine in the ratio of 35:50:15 mol were prepared as described previously (17). Twenty micrograms of protein in 300 µL of binding buffer (HEPES-buffered saline at pH 7.4, containing 2 mmol/L ionic calcium [Ca2+]) was added to 100 µL LMVs (35–40 nmol PS). The mixture was incubated for 10 minutes at 25°C, with gentle orbital mixing, before being centrifuged for 5 minutes at 12 000g at room temperature. The supernatant was discarded, and the LMV pellet was washed in binding buffer before being dissolved in 50 µL of sample preparation buffer (NuPage LDS; Invitrogen, Grand Island, NY). A 15-µL sample was analyzed by using sodium dodecyl sulfate polyacrylamide gel electrophoresis with a 4%–12% gradient bis-Tris gel (NuPage; Invitrogen). Measurements were performed by one author (A.S.K.).
Flow Cytometry
Binding of the contrast agents to dying cells was analyzed by using flow cytometry. Sixteen hours after induction of apoptosis in EL-4 murine lymphoma cells, by using 15 µmol/L etoposide (Eposin; PCH Pharmachemie, Haarlem, the Netherlands) (17), cell pellets (106 cells) were resuspended in 100 µL of a solution of either GST-C2A conjugated with FITC (GST-C2A-FITC) or GST-C2A conjugated with FITC and p-SCN-Bn-DTPA and loaded with Gd3+ (GST-C2A-Gd-FITC) or the PS-inactive site-directed mutants, PS-inactive GST-C2A-FITC or PS-inactive GST-C2A-Gd-FITC (20 µg/mL), in binding buffer. Two microliters of propidium iodide (100 µg/mL) was also added to detect necrotic cells, and the cells were analyzed in a cytometer (Cyan ADP-MLE; Dako, Ely, Cambridgeshire, England); 20 000 cells were counted per event. Measurements were performed and results were evaluated by two authors (A.S.K. and A.A.N.).
MR Imaging in Vitro
All MR imaging experiments were performed at 9.4 T with a vertical 89-mm bore magnet interfaced to a console (Inova; Varian, Palo Alto, Calif) by using a 45-mm-diameter coil (Millipede; Varian).
EL-4 cells (107) that had been treated for 16 hours with 15 µmol/L etoposide were incubated with increasing concentrations (0–66 µmol/L) of either PS-active GST-C2A conjugated with p-SCN-Bn-DTPA and loaded with Gd3+ (GST-C2A-Gd) or PS-inactive GST-C2A-Gd in binding buffer at 20°C for 10 minutes. T1 maps (T1 determinations for each voxel in the image) were acquired from cells that had been washed to remove unbound contrast agent. Transverse inversion-recovery fast low-angle shot MR images (repetition time msec/echo time msec/inversion time msec, 5.5/2/50–15 000; inversion times, 15; data matrix, 128 x 64; field of view, 35 x 35 mm; center-out data collection; transients per increment, two; section thickness, 1 mm; relaxation delay, 15 seconds) were obtained. Untreated cells were used as controls.
MR Imaging in Vivo
EL-4 cells (5 x 106) were injected subcutaneously in the flanks of 28 female C57BL/6 mice (Charles River, Wilmington, Mass) that were 6–8 weeks old. When tumor volume reached approximately 1.3 cm3 (typically 2 weeks after implantation), animals were treated with a single intraperitoneal injection of either etoposide (67 mg/kg) and cyclophosphamide monohydrate (Sigma-Aldrich) (100 mg/kg) or drug solvent alone. Sixteen hours later, MR images were acquired before contrast agent injection. Images were obtained at 10 minutes, 4 hours, and 24 hours after contrast agent injection.
Four groups of animals were compared: drug-treated animals injected with PS-active (n = 10) and PS-inactive (n = 6) contrast agents and untreated animals injected with PS-active (n = 6) and PS-inactive (n = 6) contrast agents, hereafter referred to as treated–PS-active, treated–PS-inactive, untreated–PS-active, and untreated–PS-inactive groups, respectively. Animals were assigned randomly to these groups. Initially, the concentrations of the PS-active and PS-inactive contrast agents were adjusted so that the relaxation rates of the injected contrast agents were the same. A 200-µL solution of 0.5 mmol/L (equivalent to 200 mg/kg) PS-active contrast agent (GST-C2A-Gd or GST-C2A-Gd-FITC) (r1 = 60 [mmol/L]–1 · sec–1) was injected intravenously into both treated (n = 8) and untreated (n = 4) animals. For the PS-inactive contrast agent, 200 µL of a 0.3 mmol/L solution (125 mg/kg) of PS-inactive GST-C2A-Gd or PS-inactive GST-C2A-Gd-FITC (r1 = 100 [mmol/L]–1 · sec–1) was injected into both drug-treated (n = 4) and untreated (n = 4) animals.
To determine whether the difference in specific relaxivity values of the PS-active and PS-inactive conjugates had any effect, further experiments were conducted with PS-active GST-C2A-Gd-FITC and PS-inactive GST-C2A-Gd-FITC, which had the same relaxivity and therefore could be administered at the same concentration (200 µL, 0.5 mmol/L, r1 = 50 [mmol/L]–1 · sec–1) in the treated–PS-active group (n = 2), treated–PS-inactive group (n = 2), untreated–PS-active group (n = 2), and untreated–PS-inactive group (n = 2). Animals were anesthetized with intraperitoneal injections of two medications (Hypnorm, Jansen Pharmaceuticals, High Wycombe, Buckinghamshire, England; Hypnovel, Roche, Welwyn Garden City, Hertfordshire, England) and a dextrose-saline mixture (ratio, 4%:0.18%) in a 5:4:31 ratio (10 mL per kilogram of body weight).
Transverse fat-saturated T1-weighted spin-echo (repetition time msec/echo time msec, 450/8; field of view, 35 x 35 mm; data matrix, 256 x 256; section thickness, 1 mm; number of sections, 11; gap between sections, none) and inversion-recovery fast low-angle shot (5.5/2/50, 200, 400, 800, 1200, 1600, 2000, 2500, 3000, 5000, 10 000; inversion times, 11; delay between images, 10 seconds; transients per increment, two; data matrix, 128 x 64; center-out data collection) MR images were collected before contrast agent injection, immediately after contrast agent injection, and at 4 and 24 hours after contrast agent injection. T1 relaxation time maps were calculated from inversion-recovery data, and regions of interest that included the tumor and muscle were analyzed. The animals were sacrificed at the end of the experiment, and the tumors were excised for histologic evaluation.
Blood clearance of PS-active GST-C2A-Gd and PS-inactive GST-C2A-Gd was estimated by using aortic T1 measurements (inversion-recovery fast low-angle shot MR imaging, with parameters as indicated before). The aorta was identified from a transverse T1-weighted gradient-echo MR image (40/3; field of view, 25 x 25 mm; data matrix, 256 x 128; section thickness, 2 mm). Three tumor-bearing, drug-treated animals were injected with PS-inactive GST-C2A-Gd (200 µL, 0.5 mmol/L, r1 = 50 [mmol/L]–1 · sec–1), and two animals were injected with PS-active GST-C2A-Gd (200 µL, 0.5 mmol/L, r1 = 50 [mmol/L]–1 · sec–1). The clearance was followed for 3 hours after injection of the contrast agents. All MR measurements were performed by three authors (M.I.K., A.S.K., and D.E.H.) and evaluated by one author (M.I.K.).
Tumor Histologic Evaluation
Tumors were fixed in 10% formalin and embedded in paraffin. Five-micrometer sections were viewed with a fluorescence microscope (Axiovert 100 TV; Zeiss, Welwyn Garden City, Hertfordshire, England). Then, they were immediately stained by using an apoptosis detection system (DeadEnd colorimetric terminal deoxynucleotidyl transferase biotin dUTP nick end labeling [TUNEL]; Promega, Southampton, Hampshire, England) and counterstained with hematoxylin. Experiments were performed by one author (D.E.H., with 17 years of experience with these techniques).
Statistical Analysis
Data are shown as means ± standard errors of the mean. Statistical analysis was performed on data obtained from the treated–PS-active, treated–PS-inactive, untreated–PS-active, and untreated–PS-inactive groups in vivo at 24 hours. One-way analysis of variance was used to determine whether there was a difference in means. Post hoc analysis was performed by using the Dunnett method to determine whether the difference in means between the treated–PS-active group and the other groups was significant. A difference with P < .05 was regarded as significant. All analyses were performed by using spreadsheet software (Microsoft Excel, version 10, 2001; Microsoft, Bothell, Wash).
|
RESULTS
|
|---|
Binding of GST-C2A-Gd to PS
The mean dissociation constant of GST-C2A was 95 nmol/L ± 9 (standard error of the mean) (n = 3), and modification resulted in a decreased affinity for PS (Fig 1). Both GST-C2A-Gd and GST-C2A-Gd-FITC had comparable mean dissociation constants of 333 nmol/L ± 85 (n = 3) and 333 nmol/L ± 45 (n = 3), respectively.
View larger version (29K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1: Analysis with electrophoresis of the binding of PS-active and PS-inactive contrast agents to LMVs containing PS. Lane 1 = molecular weight markers (36.5–55.4 kDa), lane 2 = unmodified PS-active GST-C2A. In lanes 3–6, the specified proteins were incubated with LMVs, the LMVs washed, and the proteins released and run on the gel. Lane 3 = unmodified PS-active GST-C2A, lane 4 = PS-active GST-C2A-Gd, lane 5 = unmodified PS-inactive GST-C2A, and lane 6 = PS-inactive GST-C2A-Gd.
|
|
The additional mass, resulting from labeling with the Gd3+ chelate, was evident from the slower migration of GST-C2A-Gd in lane 4 in Figure 1. This analysis demonstrated that the conjugate retained approximately 60% of the activity of the unmodified protein and that unmodified PS-inactive GST-C2A showed reduced binding to PS (10% of unmodified GST-C2A), with the PS-inactive conjugate (PS-inactive GST-C2A-Gd) showing no detectable binding.
Analysis of Contrast Agent Binding to Cells by Using Flow Cytometry
GST-C2A-FITC bound to late apoptotic and necrotic cells and early apoptotic cells, with little binding to viable cells (Fig 2a). There was little detectable binding of PS-inactive GST-C2A-FITC to any cell population (Fig 2b). Addition of Gd3+ chelates to the native protein (GST-C2A-Gd-FITC) resulted in a reduction in binding to early apoptotic cells, although the protein still bound late apoptotic and necrotic cells (Fig 2c), with minimal binding to viable cells. Median fluorescein fluorescence intensity was 365 for dying cells compared with 28 for viable cells. Addition of Gd3+ chelates to the site-directed PS-inactive mutant (inactive GST-C2A-Gd-FITC) resulted in binding of the protein to late apoptotic and necrotic cells (Fig 2d), although this binding was less than that observed with GST-C2A-Gd-FITC. Median fluorescein fluorescence intensity was 159 for dying cells and 19 for viable cells. Both PS-active GST-C2A-Gd-FITC and PS-inactive GST-C2A-Gd-FITC had an FITC-protein ratio of approximately 0.5. Thus, both Gd3+ chelate–labeled proteins showed binding to late apoptotic and necrotic cells, but the native PS-active protein showed increased binding to this fraction and also showed binding to early apoptotic cells. Neither protein bound to viable cells.
View larger version (65K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2a: Flow cytometric analysis of contrast agent binding to etoposide-treated EL-4 cells. (a–d) Scatterplots show log fluorescein fluorescence emission (x-axis) versus log propidium iodide fluorescence emission (y-axis). (a, c, and d) Quadrants (R1–R4) are defined on the basis of fluorescence intensity values; R3 contained viable cells, R2 contained apoptotic cells, and R4 contained late-stage apoptotic and necrotic cells. In b, where there was no detectable binding of PS-inactive GST-C2A-FITC, R3 contained viable and apoptotic cells and R4 contained late-stage apoptotic and necrotic cells. Cells were incubated with 20 µg/mL of (a) PS-active GST-C2A-FITC, (b) PS-inactive GST-C2A-FITC, (c) PS-active GST-C2A-Gd-FITC, and (d) PS-inactive GST-C2A-Gd-FITC.
|
|
View larger version (41K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2b: Flow cytometric analysis of contrast agent binding to etoposide-treated EL-4 cells. (a–d) Scatterplots show log fluorescein fluorescence emission (x-axis) versus log propidium iodide fluorescence emission (y-axis). (a, c, and d) Quadrants (R1–R4) are defined on the basis of fluorescence intensity values; R3 contained viable cells, R2 contained apoptotic cells, and R4 contained late-stage apoptotic and necrotic cells. In b, where there was no detectable binding of PS-inactive GST-C2A-FITC, R3 contained viable and apoptotic cells and R4 contained late-stage apoptotic and necrotic cells. Cells were incubated with 20 µg/mL of (a) PS-active GST-C2A-FITC, (b) PS-inactive GST-C2A-FITC, (c) PS-active GST-C2A-Gd-FITC, and (d) PS-inactive GST-C2A-Gd-FITC.
|
|
View larger version (68K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2c: Flow cytometric analysis of contrast agent binding to etoposide-treated EL-4 cells. (a–d) Scatterplots show log fluorescein fluorescence emission (x-axis) versus log propidium iodide fluorescence emission (y-axis). (a, c, and d) Quadrants (R1–R4) are defined on the basis of fluorescence intensity values; R3 contained viable cells, R2 contained apoptotic cells, and R4 contained late-stage apoptotic and necrotic cells. In b, where there was no detectable binding of PS-inactive GST-C2A-FITC, R3 contained viable and apoptotic cells and R4 contained late-stage apoptotic and necrotic cells. Cells were incubated with 20 µg/mL of (a) PS-active GST-C2A-FITC, (b) PS-inactive GST-C2A-FITC, (c) PS-active GST-C2A-Gd-FITC, and (d) PS-inactive GST-C2A-Gd-FITC.
|
|
View larger version (59K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2d: Flow cytometric analysis of contrast agent binding to etoposide-treated EL-4 cells. (a–d) Scatterplots show log fluorescein fluorescence emission (x-axis) versus log propidium iodide fluorescence emission (y-axis). (a, c, and d) Quadrants (R1–R4) are defined on the basis of fluorescence intensity values; R3 contained viable cells, R2 contained apoptotic cells, and R4 contained late-stage apoptotic and necrotic cells. In b, where there was no detectable binding of PS-inactive GST-C2A-FITC, R3 contained viable and apoptotic cells and R4 contained late-stage apoptotic and necrotic cells. Cells were incubated with 20 µg/mL of (a) PS-active GST-C2A-FITC, (b) PS-inactive GST-C2A-FITC, (c) PS-active GST-C2A-Gd-FITC, and (d) PS-inactive GST-C2A-Gd-FITC.
|
|
Analysis of Contrast Agent Binding to Cells by Using MR Imaging
Pellets of dying EL-4 cells, which had been incubated with increasing concentrations of PS-active GST-C2A-Gd and PS-inactive GST-C2A-Gd, showed progressive decreases in T1 (Fig 3). A greater R1 (1/T1) change, which is proportional to the bound concentration of the contrast agent, was observed with the native PS-active protein at any given concentration (Fig 3b). Neither contrast agent, PS-active GST-C2A-Gd or PS-inactive GST-C2A-Gd, showed much binding to viable cells (Fig 3c, point C).
Detection of Cell Death in Vivo by Using MR Imaging
Shorter tumor T1 values were observed in drug-treated animals 24 hours after injection with the PS-active contrast agent (PS-active GST-C2A-Gd) (Fig 4, A) compared with values in the control animals. The control animals included drug-treated animals injected with the PS-inactive contrast agent (PS-inactive GST-C2A-Gd) (Fig 4, B) and untreated animals injected with PS-active (Fig 4, C) and PS-inactive contrast agents (Fig 4, D). The PS-inactive contrast agent shows lesser binding to dying cells. The relaxation data were normalized to the postinjection R1 value obtained at 10 minutes (referred to hereafter as the R1 ratio) (Fig 5). In all experiments, the accumulation of PS-active contrast agent in the tumors of drug-treated animals was significantly greater than it was in the tumors of the control animals. Because injection of PS-active and PS-inactive contrast agents, which had been matched for relaxation rate or for both concentration and relaxation rate, produced similar results, the pooled data set was used to calculate mean R1 ratios. For example, the mean R1 ratio for treated animals injected with PS-inactive contrast agent matched for relaxation rate was 0.99 ± 0.11 compared with 1.08 for animals injected with PS-inactive contrast agent matched for concentration and relaxation rate. Similarly, the mean ratio was 0.95 ± 0.09 for untreated animals injected with the first preparation and 0.98 for animals injected with the second preparation. The PS-active contrast agent was always used at the same concentration and approximately the same relaxivity.
View larger version (49K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4: Transverse T1 maps of drug-treated and untreated tumors in animals injected with PS-active (GST-C2A-Gd) and PS-inactive (GST-C2A-Gd) contrast agents. Color scale indicates T1 values for image voxels. In this example, contrast agents were matched for relaxation rate. Images were acquired immediately before injection of contrast agent (a T1 map acquired from a tumor before injection is shown on the left-hand side) and at 24 hours after injection. Reference capillary was placed adjacent to the tumors, which were implanted on lower areas of backs of animals. Position of the tumor is indicated on the gray-scale image. A, Drug-treated tumor in animal injected with PS-active GST-C2A-Gd (TA). B, Drug-treated tumor in animal injected with PS-inactive GST-C2A-Gd (TI). C, Untreated tumor in animal injected with PS-active GST-C2A-Gd (UA). D, Untreated tumor in animal injected with PS-inactive GST-C2A-Gd (UI). Drug-treated tumor in animal injected with PS-active contrast agent shows greater accumulation at 24 hours after injection (A).
|
|
In all animals, the mean muscle R1 ratios recovered to their preinjection values (Fig 5a, 5b), indicating good tissue clearance of the contrast agents. The mean tumor R1 ratio in the treated–PS-active group at 24 hours after injection was 1.19 ± 0.04, which was significantly higher (P < .05) than that of any of the control animals (treated–PS-inactive group: mean, 1.02 ± 0.04; untreated–PS-active group: mean, 1.00 ± 0.03; untreated–PS-inactive group: mean, 0.96 ± 0.03), indicating greater accumulation of the PS-active contrast agent in the drug-treated tumors. In the treated tumors injected with the PS-inactive contrast agent, there was retention of the contrast agent at 24 hours after injection, but the amount retained was less than that observed with the PS-active contrast agent, a finding that is consistent with the lesser binding of the PS-inactive contrast agent to dying cells. The mean blood half-life for the PS-active contrast agent was 201 minutes ± 7 and that for the PS-inactive contrast agent was 186 minutes ± 7. The contrast agents showed evidence of renal clearance since substantial shortening of the bladder T1 was observed in some animals at 24 hours.
Tumor Histologic Evaluation Findings
Paraffin-embedded tumor sections obtained from animals that had been injected with GST-C2A-Gd-FITC were examined by using fluorescence microscopy (Fig 6, right). The same sections were then stained by using the TUNEL method and counterstained with hematoxylin (Fig 6, left). Areas of cell death could be readily visualized in the TUNEL-stained sections and differentiated from viable tissue in these tumors. GST-C2A-Gd-FITC appeared to localize to these areas of cell death.
View larger version (85K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6: Left: TUNEL-stained section from drug-treated tumor in animal injected with GST-C2A-Gd-FITC. TUNEL-positive nuclei are stained brown (bar = 200 µm). Tissue was counterstained with hematoxylin. Inset shows same area at higher magnification (bar = 50 µm). Right: Fluorescence image from same section (bar = 200 µm). Rectangle = location of inset shown in bottom left-hand corner of left image.
|
|
|
DISCUSSION
|
|---|
A Gd3+-based targeted MR imaging contrast agent that can be used to detect tumor cell death in vivo by binding to exposed PS has been described. The native, PS-active Gd3+ chelate–labeled protein clearly showed greater accumulation in treated tumors when compared with the PS-inactive mutant protein, a finding that is consistent with the increased binding of the PS-active contrast agent to dying cells observed in vitro. The differences in retention of the PS-active and PS-inactive contrast agents cannot be explained by differences in their pharmacokinetics because the clearance of both PS-active GST-C2A-Gd and PS-inactive GST-C2A-Gd was similar.
The estimated dissociation constant of GST-C2A for PS was similar to that reported previously (19), although it was higher than that reported for C2A (17 nmol/L) (17) and annexin A5 (0.1 nmol/L) (20). Attachment of between 10 and 14 Gd3+-DTPA chelates to GST-C2A decreased its affinity for PS. Modification of annexin A5 similarly results in loss of activity, where only 1.6 modifications per mole of protein were shown to result in significant loss of PS binding activity (21).
Comparison of contrast agent clearance from underlying muscle and tumor tissue showed some evidence of nonspecific trapping of both contrast agents in the tumors after 24 hours. This reflects a limitation of the tumor model used in our study, where its relatively rapid growth and treatment response precluded measurements beyond 24 hours, when it is expected that clearance of this trapped material should become apparent. Histologic analysis of treated tumors indicated that there was 20%–30% cell death, which is comparable with levels found in some human tumors after treatment (22,23). However, the relatively small change in T1 may limit detection of the contrast agent in the clinic. Detection could be improved by increasing the Gd3+ payload, for example, by covalently attaching C2A to a Gd3+ chelate–loaded polyamidoamine, known as PAMAM, dendrimer (24,25). The Gd3+ concentrations administered in these experiments, 50–75 µmol/kg, are lower than those used clinically with low–molecular-weight contrast agents such as gadopentetate dimeglumine (Schering, Berlin, Germany) and gadoterate meglumine (Dotarem; Guerbet, Roissy, France), where a typical dose is approximately 0.1–0.3 mmol/kg body weight (26). The magnetic field strength used here is much higher than the strengths used clinically, resulting in higher signal-to-noise ratios, and, thus, potentially limiting the relevance of the results of this study to application in the clinic. However, macromolecular Gd3+-based contrast agents can have substantially higher relaxivity values at lower magnetic field strengths (25).
The targeted Gd3+-based contrast agent described here for detecting apoptosis has substantial advantages compared with the SPIO-based contrast agent described previously (15). The contrast agent generates positive contrast, making it easier to detect on the irregular and intrinsically heterogeneous images obtained from tumors. It is much smaller (3–6 nm as compared with 50–100 nm) than SPIO-based contrast agent (12,15,17) and, thus, should penetrate and exit the tumor interstitium more easily, generating improved contrast.
The targeted contrast agent described here has been demonstrated to have the capability for detection of tumor cell death in vivo, and future derivatives of it could potentially be used clinically to assess early tumor responses to treatment.
|
ADVANCES IN KNOWLEDGE
|
|---|
- Conjugation of ionic gadolinium (Gd3+) chelates to a protein that binds to the phosphatidylserine exposed by dying cells has produced a targeted contrast agent that enables the noninvasive MR imaging detection of cell death in tumors after treatment with a chemotherapeutic agent.
- A site-directed mutant of the protein showed no detectable binding to apoptotic cells and much weaker binding to necrotic cells than the labeled native protein.
|
IMPLICATION FOR PATIENT CARE
|
|---|
- Application in the clinic of a Gd3+-based MR imaging contrast agent, similar to the one described here, may have advantages over existing noninvasive methods for detection of tumor cell death after therapy.
|
FOOTNOTES
|
|---|
Abbreviations: DTPA = diethylenetriaminepentaacetic acid • FITC = fluorescein isothiocyanate • GST = glutathione-S-transferase • GST-C2A = GST–C2A fusion protein • GST-C2A-FITC = GST-C2A conjugated with FITC • GST-C2A-Gd = GST-C2A conjugated with p-SCN-Bn-DTPA and loaded with Gd3+ • GST-C2A-Gd-FITC = GST-C2A conjugated with FITC and p-SCN-Bn-DTPA and loaded with Gd3+ • LMV = large multilamellar vesicles • PS = phosphatidylserine • SPIO = superparamagnetic iron oxide • TUNEL = terminal deoxynucleotidyl transferase biotin dUTP nick end labeling
Author contributions: Guarantor of integrity of entire study, K.M.B.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, all authors; experimental studies, A.S.K., A.A.N., M.M.d.B., D.E.H., B.D., M.I.K.; statistical analysis, A.S.K., A.A.N., M.I.K.; and manuscript editing, A.S.K., A.A.N., M.M.d.B., B.D., M.I.K., K.M.B.
Authors stated no financial relationship to disclose.
|
References
|
|---|
- Therasse P, Arbuck SG, Eisenhauer EA, et al. New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst 2000;92:205–216.
- Meyn RE, Stephens LC, Hunter NR, Milas L. Apoptosis in murine tumors treated with chemotherapy agents. Anticancer Drugs 1995;6:443–450. [Medline]
- Ellis PA, Smith IE, McCarthy K, Detre S, Salter J, Dowsett M. Preoperative chemotherapy induces apoptosis in early breast cancer. Lancet 1997;349:849. [CrossRef][Medline]
- Dubray B, Breton C, Delic J, et al. In vitro radiation-induced apoptosis and early response to low-dose radiotherapy in non-Hodgkin's lymphomas. Radiother Oncol 1998;46:185–191. [CrossRef][Medline]
- Chang J, Ormerod M, Powles TJ, Allred DC, Ashley SE, Dowsett M. Apoptosis and proliferation as predictors of chemotherapy response in patients with breast carcinoma. Cancer 2000;89:2145–2152. [CrossRef][Medline]
- Fadok VA, Bratton DL, Frasch SC, Warner ML, Henson PM. The role of phosphatidylserine in recognition of apoptotic cells by phagocytes. Cell Death Differ 1998;5:551–562. [CrossRef][Medline]
- Savill J, Fadok V. Corpse clearance defines the meaning of cell death. Nature 2000;407:784–788. [CrossRef][Medline]
- Vermes I, Haanen C, Steffens-Nakken H, Reutelingsperger C. A novel assay for apoptosis: flow cytometric detection of phosphatidylserine expression on early apoptotic cells using fluorescein labelled annexin V. J Immunol Methods 1995;184:39–51. [CrossRef][Medline]
- Blankenberg FG, Katsikis PD, Tait JF, et al. In vivo detection and imaging of phosphatidylserine expression during programmed cell death. Proc Natl Acad Sci U S A 1998;95:6349–6354. [Abstract/Free Full Text]
- Lahorte CM, Vanderheyden JL, Steinmetz N, Van de Wiele C, Dierckx RA, Slegers G. Apoptosis-detecting radioligands: current state of the art and future perspectives. Eur J Nucl Med Mol Imaging 2004;31:887–919. [CrossRef][Medline]
- Petrovsky A, Schellenberger E, Josephson L, Weissleder R, Bogdanov A Jr. Near-infrared fluorescent imaging of tumor apoptosis. Cancer Res 2003;63:1936–1942. [Abstract/Free Full Text]
- Schellenberger EA, Bogdanov A Jr, Hogemann D, Tait J, Weissleder R, Josephson L. Annexin V-CLIO: a nanoparticle for detecting apoptosis by MRI. Mol Imaging 2002;1:102–107. [CrossRef][Medline]
- Schellenberger EA, Bogdanov A Jr, Petrovsky A, Ntziachristos V, Weissleder R, Josephson L. Optical imaging of apoptosis as a biomarker of tumor response to chemotherapy. Neoplasia 2003;5:187–192. [Medline]
- Davletov BA, Sudhof TC. A single C2 domain from synaptotagmin I is sufficient for high affinity Ca2+/phospholipid binding. J Biol Chem 1993;268:26386–26390. [Abstract/Free Full Text]
- Zhao M, Beauregard DA, Loizou L, Davletov B, Brindle KM. Non-invasive detection of apoptosis using magnetic resonance imaging and a targeted contrast agent. Nat Med 2001;7:1241–1244. [CrossRef][Medline]
- Zhao M, Zhu X, Ji S, et al. 99mTc-labeled C2A domain of synaptotagmin I as a target-specific molecular probe for noninvasive imaging of acute myocardial infarction. J Nucl Med 2006;47:1367–1374. [Abstract/Free Full Text]
- Jung HI, Kettunen MI, Davletov B, Brindle KM. Detection of apoptosis using the C2A domain of synaptotagmin I. Bioconjug Chem 2004;15:983–987. [CrossRef][Medline]
- Niemi P, Koskinen S, Reisto T. Tissue relaxation enhancement after intravenous administration of (ITCB-DTPA)-gadolinium conjugated albumin, an intravascular magnetic resonance imaging contrast agent. Invest Radiol 1991;26:674–680. [CrossRef][Medline]
- Lu YJ, He Y, Sui SF. Inositol hexakisphosphate (InsP6) can weaken the Ca2+-dependent membrane binding of C2AB domain of synaptotagmin I. FEBS Lett 2002;527:22–26. [CrossRef][Medline]
- Tait JF, Gibson D, Fujikawa K. Phospholipid binding properties of human placental anticoagulant protein-I, a member of the lipocortin family. J Biol Chem 1989;264:7944–7949. [Abstract/Free Full Text]
- Schellenberger EA, Weissleder R, Josephson L. Optimal modification of annexin V with fluorescent dyes. Chembiochem 2004;5:271–274. [CrossRef][Medline]
- Jansen B, Wacheck V, Heere-Ress E, et al. Chemosensitisation of malignant melanoma by BCL2 antisense therapy. Lancet 2000;356:1728–1733. [CrossRef][Medline]
- Symmans WF, Volm MD, Shapiro RL, et al. Paclitaxel-induced apoptosis and mitotic arrest assessed by serial fine-needle aspiration: implications for early prediction of breast cancer response to neoadjuvant treatment. Clin Cancer Res 2000;6:4610–4617. [Abstract/Free Full Text]
- Wiener EC, Brechbiel MW, Brothers H, et al. Dendrimer-based metal chelates: a new class of magnetic resonance imaging contrast agents. Magn Reson Med 1994;31:1–8. [Medline]
- Kobayashi H, Brechbiel MW. Dendrimer-based macromolecular MRI contrast agents: characteristics and application. Mol Imaging 2003;2:1–10. [CrossRef][Medline]
- Brücher E, Sherry AD. Stability and toxicity of contrast agents. In: Merbach AE, Tóth É, eds. The chemistry of contrast agents in medical magnetic resonance imaging. Chichester, England: Wiley, 2001; 245–279.
This article has been cited by other articles:
|
|
|
|
 
J. C. Gore, T. E. Yankeelov, Todd. E. Peterson, and M. J. Avison
Molecular Imaging Without Radiopharmaceuticals?
J. Nucl. Med.,
June 1, 2009;
50(6):
999 - 1007.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
B. Banihashemi, R. Vlad, B. Debeljevic, A. Giles, M. C. Kolios, and G. J. Czarnota
Ultrasound Imaging of Apoptosis in Tumor Response: Novel Preclinical Monitoring of Photodynamic Therapy Effects
Cancer Res.,
October 15, 2008;
68(20):
8590 - 8596.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
M. Zhao, Z. Li, and S. Bugenhagen
99mTc-Labeled Duramycin as a Novel Phosphatidylethanolamine-Binding Molecular Probe
J. Nucl. Med.,
August 1, 2008;
49(8):
1345 - 1352.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
) and PS-active GST-C2A-Gd (
). Control experiments were conducted with viable nondrug-treated cells (C). Cell samples (106 cells in 50 µL) were placed in Eppendorf tubes, the cells were concentrated into pellets by using centrifugation, and the imaging section was placed through the cell pellets.
) tumors in animals injected with active GST-C2A-Gd or GST-C2A-Gd-FITC. There was a significant difference (** = P < .01, analysis of variance, Dunnett post hoc analysis) after 24 hours in the R1 ratio of tumors in the drug-treated group (mean, 1.19 ± 0.04). Underlying muscle tissue in drug-treated (
) and untreated (
) animals injected with PS-active contrast agent showed evidence of clearance of the contrast agent. R1 ratios are mean ± standard error of the mean. (b) R1 ratio for drug-treated tumors in animals injected with PS-active GST-C2A-Gd-FITC (
