Bioconjugate Chem. 2010, 21, 531–536
531
Evaluation of Bifunctional Chelates for the Development of Gallium-Based Radiopharmaceuticals Cara L. Ferreira,*,† Eric Lamsa,† Michael Woods,† Yin Duan,‡ Pasan Fernando,‡ Corinne Bensimon,‡ Myra Kordos,‡ Katharina Guenther,§ Paul Jurek,| and Garry E. Kiefer| MDS Nordion, 4004 Wesbrook Mall, Vancouver, British Columbia, Canada V6T 2A3, Ottawa Heart Institute, Ottawa, Ontario, Canada, MDS Nordion, Kanata, Ontario, Canada, and Macrocyclics, Dallas, Texas. Received October 9, 2009; Revised Manuscript Received January 4, 2010
Ga radioisotopes, including the generator-produced positron-emitting isotope 68Ga (t1/2 ) 68 min), are of increasing interest for the development of new radiopharmaceuticals. Bifunctional chelates (BFCs) that can be efficiently radiolabeled with Ga to yield complexes with good in ViVo stability are needed. To this end, we undertook a systematic comparison of four BFCs containing different chelating moieties: two novel BFCs, p-NO2-Bn-Oxo (1-oxa-4,7,10-triazacyclododecane-4,7,10-triacetic acid) and p-NO2-Bn-PCTA (3,6,9,15-tetraazabicyclo [9.3.1]pentadeca-1(15),11,13-triene-3,6,9-triacetic acid), and two more commonly used BFCs, p-NO2-Bn-DOTA (1,4,7,10tetraazacyclododecane-1,4,7,10-tetraacetic acid) and p-NO2-Bn-NOTA (1,4,7-triazacyclononane-1,4,7-triacetic acid). Each BFC was compared with respect to radiolabeling conditions, radiochemical yield, stability, and in vivo clearance properties. p-NO2-Bn-PCTA, p-NO2-Bn-Oxo, and p-NO2-Bn-NOTA were all more efficiently radiolabeled with Ga compared to p-NO2-Bn-DOTA. p-NO2-Bn-DOTA required longer reaction time, higher concentrations of BFC, or heating to obtain equivalent radiochemical yields. Better stability was observed for p-NO2-Bn-NOTA and p-NO2-Bn-PCTA compared to p-NO2-Bn-DOTA and p-NO2-Bn-Oxo, especially with respect to transmetalation to transferrin. Ga-radiolabled p-NO2-Bn-Oxo was found to be kinetically labile and therefore unstable in vivo. Ga-radiolabeled p-NO2-Bn-NOTA and p-NO2-Bn-PCTA were relatively inert, while Ga-radiolabeled p-NO2-BnDOTA had intermediate stability, losing >20% of Ga in less than one hour when incubated with apo-transferrin. Similar stability differences were seen when incubating at pH 2. In vivo PET imaging and biodistribution studies in mice showed that 68Ga-radiolabeled p-NO2-Bn-PCTA, p-NO2-Bn-NOTA, and p-NO2-Bn-DOTA all cleared through the kidneys. While there was no statistical difference in the biodistribution results of 68Ga-radiolabeled p-NO2-Bn-PCTA and p-NO2-Bn-DOTA, 68Ga-radiolabeled p-NO2-Bn-NOTA cleared more rapidly from blood and muscle tissue but retained at up to 5 times higher activity in the kidneys.
INTRODUCTION Gallium radioisotopes have long been of interest in nuclear medicine. 67Ga is currently used for infection and tumor imaging (1, 2). Another gallium isotope gaining the attention of the nuclear imaging community is 68Ga (3-5). 68Ga is a generator-produced positron-emitter with a 68 min half-life. There are only a few generator-produced isotopes applicable to positron emission tomography (PET) imaging, and most have short half-lives limiting their applications due to the time requirements for both radiolabeling and target uptake in vivo. The wide use of single photon emission computed tomography (SPECT) imaging can be partially attributed to the convenience of the Tc-99 m generator and the simplicity of the kit chemistry. Despite having higher resolution, PET has had slower adoption because of the need for a nearby cyclotron to produce F-18 and the more complex chemistry needed to incorporate F-18 into the radiopharmaceutical. The availability of a 68Ga generator and kit chemistry to produce 68Ga-based molecular imaging agents could have a significant impact on PET (4). For 68Ga to be effectively used in molecular imaging, agents must be developed with applicable radiochemistry and appropri* Cara L. Ferreira;
[email protected], phone 1-604-2288952, fax 1-604-228-5990. † MDS Nordion, British Columbia. ‡ Ottawa Heart Institute. § MDS Nordion, Ontario. | Macrocyclics.
ate biological properties. Due to the short half-life, the radiolabel incorporation needs to be efficient, preferably under mild conditions, and not require subsequent purification. Radiolabeling that yields an injectable radiopharmaceutical with high specific activity after simply adding the 68Ga generator eluant directly to a reagent containing vial would be ideal. The agent also needs to be stable in ViVo and clear background tissues quickly to allow imaging within the first few hours after injection. A factor that influences the radiochemistry as well as the biological properties of the agent is the chemical method of incorporating 68Ga. Typically, a bifunctional chelate (BFC) is used to stably chelate the metal and link the metal to a target vector, such as a peptide (6). Choosing the right BFC can greatly affect the radiolabeling kinetics and conditions, increase the resulting compound’s stability, and influence, to some degree, the pharmacokinetics. The most commonly used BFC for 68Ga is DOTA (1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid), but the radiochemistry typically requires conventional or microwave heating to obtain adequate yields and specific activities quickly (7-11). More recently, NOTA (1,4,7-triazacyclononane-1,4,7-triacetic acid) and derivatives have been radiolabeled efficiently with 68Ga at lower temperatures (12, 13) and even at ambient temperature (14, 15). Previously, we have examined two novel bifunctional chelates, p-NO2-Bn-PCTA and p-NO2-Bn-Oxo, for the development of Cu-64 based radiopharmaceuticals (16). In this work, we examined the same two BFCs for their potential use in
10.1021/bc900443a 2010 American Chemical Society Published on Web 02/22/2010
532 Bioconjugate Chem., Vol. 21, No. 3, 2010
developing gallium-radiolabeled nuclear imaging agents. These BFCs each contain a 12-membered macrocyclic chelate, p-NO2Bn-Oxo (1-oxa-4,7,10-triazacyclododecane-4,7,10-triacetic acid), p-NO2-Bn-PCTA (3,6,9,15-tetraazabicyclo [9.3.1]pentadeca1(15),11,13-triene-3,6,9-triacetic acid), and a p-nitrobenzyl substituent, which can be modified for attachment of a targeting vector. To determine whether p-NO2-Bn-PCTA and p-NO2-BnOxo offered an advantage over the currently used BFCs for Gabased radiopharmaceutical development, direct comparison of the novel BFCs was made to the analogous BFCs containing the DOTA and NOTA chelating moieties. The BFCs were examined with respect to radiochemistry and stability with both 67 Ga and 68Ga. The pharmacokinetics of the Ga-radiolabeled BFCs was also compared through biodistribution and imaging studies in mice.
EXPERIMENTAL SECTION Materials and Methods. All solvents and reagents were used as received unless otherwise noted. The BFCs, p-NO2-BnDOTA, p-NO2-Bn-NOTA, p-NO2-Bn-Oxo, and p-NO2-BnPCTA, were acquired from Macrocyclics Inc. (Texas). 67Ga and 68 Ga were obtained as 0.1 N HCl solutions (MDS Nordion Inc.). Water was deionized using a Milli-Q biocel A-10 water purification system. Reaction vials were acid-washed to remove impurities and trace metals. The HPLC system used for analysis consists of a Waters Alliance HT 2795 separation module equipped with a Raytest Gabbistar NaI detector and a Waters 996 photodiode array (PDA) detector. Analysis of radiolabeled complexes was done on a Phenomenex Hydrosynergy RP C18 4.6 × 150 mm analytical column using a binary solvent system: solvent A ) acetonitrile, solvent B ) 0.05% trifluoroacetic acid in water. Method 1 was used for p-NO2-Bn-DOTA, p-NO2-Bn-Oxo, and p-NO2-Bn-PCTA: gradient elution 5% A/95% B to 30% A/70% B over 20 min at 1.5 mL/min. Method 2 was used for p-NO2Bn-NOTA: gradient elution 5% A/95% B to 50% A/50% B over 30 min at 1.5 mL/min. Positive identification of the 67Ga and 68Ga radiolabeled BFCs was made by comparison of the retention time of the radiodetector peak and the UV detector peak associated with macroscopic scale Ga complex of the respective BFC, whose identity was verified by mass spectroscopy using positive-ion mode electrospray ionization with a Micromass Q-TOF Ultima API time-of-flight mass spectrometer by Mr. Andras Szeitz (University of British Columbia, Department of Pharmaceutical Sciences). Analysis of the stability samples was done by HPLC using the method described above or using a size-exclusion column, Phenomenex BIOSEP SEC2000 7.8 × 300 mm, eluting with 10 mM sodium bicarbonate buffer pH 6.9 at 1.0 mL/min. 67 Ga Radiolabeling Study of Reaction Kinetics. The 67Ga solution (50-100 µL, 25 mCi/mL) was added to a reaction vial containing one of the chelates, p-NO2-Bn-DOTA, p-NO2-BnNOTA, p-NO2-Bn-Oxo, or p-NO2-Bn-PCTA, in a 10 mM sodium acetate buffer at pH 4-5. The final concentration of chelate ranged 6-300 µM, and the final concentration of Ga3+ ranged 3-6 µM. The reactions were done in triplicate and were analyzed by HPLC for radiochemical yield after 5, 10, 30, and 60 min. 67 Ga Radiolabeling at Different pH Values. 67Ga (50 µL, 20 mCi/mL) was added to a solution containing one of the BFCs, p-NO2-Bn-DOTA, p-NO2-Bn-NOTA, p-NO2-Bn-Oxo, or p-NO2Bn-PCTA (50 µL, 60 µM) diluted in 1 mL of buffer (10 mM) at different pH values ranging from 2 to 6. All reactions were done in triplicate. Radiochemical yield was determined by HPLC analysis after 5 and 60 min. 68 Ga Radiolabeling Comparison and Optimization. Carrierfree 68Ga (50-400 µL, 1 mCi) was added to BFC (2-2000
Ferreira et al.
nmol) in 1 mL of 200 mM sodium acetate buffer (pH 4). Reactions proceeded at room temperature, 40, or 80 °C and were monitored by HPLC at different time points. Stability Assessment. Stability assessments were done with both 67Ga and 68Ga radiolabeled BFCs. For 67Ga, carrier-free 67 Ga (3-4 mCi) was added to one of the four BFCs (30 µM in 1.0 mL 10 mM sodium acetate buffer pH 4). After 10 min, the radiochemical yield was determined by HPLC. For 68Ga, carrierfree 68Ga (1-2 mCi) was added to one of the four BFCs (20 µM in 1.0 mL 400 mM sodium acetate buffer pH 4), and the resulting solution was heated at 80 °C for 15 min to ensure reaction completion. The radiochemical yield was confirmed by HPLC. Aliquots of the radiolabeling reactions (200 µL) were added to 800 µL of either a pH 2 buffer (0.1 mM glycine) or an apotransferrin solution (3 mg/mL, pH 6.9 10 mM sodium bicarbonate buffer). The stability solutions were incubated at 37 °C and analyzed by HPLC at different time points to determine the amount of radiolabeled BFC that remained intact. Biodistribution and Imaging in Mice. A total of 48 male CD1 (25-30 g) mice (Charles River, Canada) were used for the animal study. Each of the four different 68Ga radiolabeled BFCs prepared as described above (10-20 Ci/mmol of BFC) was injected via the lateral tail vein into 12 mice. Animals were injected under light anesthesia with isoflurane; isoflurane levels were maintained at 1.5-2%. Ten of the mice were injected with ∼10 µCi, while two of the mice were injected with ∼50 µCi and used for both imaging and the 4 h biodistribution data. The mice were sacrificed by cervical dislocation while under anesthesia. Tissues were harvested at 15 min, 1 h, and 4 h after injection, with n ) 4 for each time point and included heart, liver, kidneys, spleen, intestines, skeletal muscle (quadriceps), bone (femur), brain, blood, and urine. Tissues were weighed, and their activity was counted on a gamma counter. The tissue activity was decay-corrected and converted to %ID/g. Student t tests were used for analysis, with a p < 0.05 considered statistically significant. PET imaging was carried out on a Siemens Inveon small animal PET scanner (Siemens, Knoxville, Tennessee). The scanner was started 10 s prior to radiotracer injection in order to capture the entire input function. A bolus injection of approximately 50 µCi in 0.15 mL of the 68Ga-radiolabeled BFC was injected intravenously over a 10 s period. List-mode data were acquired for 90 min using a 350-650 keV energy window and a coincidence timing window of 3.4 ns. List-mode PET data were histogrammed two different ways (32 frames of 12 × 10 s, 3 × 60 s, and 17 × 300 s and 2 frames of 1 × 300 s and 1 × 5100 s). The data were reconstructed on a 256 × 256 image matrix with 0.29 × 0.29 × 0.80 mm3 pixel size using OSEM3D/MAP (β ) 0, OSEM3D iterations ) 2, MAP iterations ) 18) with corrections for dead-time, isotope decay, detector efficiencies, and randoms.
RESULTS Radiochemistry. Standards were prepared by adding naturally abundant gallium to each of the BFCs (Figure 1). The parent-ion weight and isotopic distribution observed in the mass spectra confirm the coordination of Ga to the BFCs; Ga-p-NO2Bn-PCTA m/z ) 581.4, 583.1; Ga-p-NO2-Bn-Oxo m/z ) 549.2, 551.2; Ga-p-NO2-Bn-DOTA m/z ) 606.1, 608.1, Ga-p-NO2Bn-NOTA m/z ) 505.1, 507.2. The natural Ga complexes were used as UV/visible detectable standards for HPLC identification of the radiolabeled compounds. Representative HPLC traces for the p-NO2-Bn-PCTA BFC with natural Ga, 67Ga, and 68Ga are shown in Figure 2. The radiolabeling yield was measured after a fixed reaction time at room temperature at several different pH values. p-NO2-
Evaluation of Bifunctional Chelates for Gallium
Bioconjugate Chem., Vol. 21, No. 3, 2010 533
Figure 3. % Ga-BFC intact after incubation in 3 mg/mL apo-transferrin at 37 °C over 4 h. *Values significantly different compared to other Ga-BFC species at the same time point (p < 0.005).
Figure 1. Bifunctional chelates evaluated for gallium radiolabeling.
Figure 2. HPLC traces for (a) 67Ga-p-NO2-Bn-PCTA (radiation detector), (b) 68Ga-p-NO2-Bn-PCTA (radiation detector), (c) Ga-p-NO2Bn-PCTA (UV/visible detector, λ ) 280 nm).
Bn-DOTA was most sensitive to variation in pH, with optimal radiochemical yields at pH 4, 99% after 60 min, and yields dropping to less than 80% at pH 3 and 5. p-NO2-Bn-PCTA, p-NO2-Bn-Oxo, and p-NO2-Bn-NOTA were radiolabeled efficiently with yields >95% between pH 3 and pH 5. All BFCs showed inferior radiolabeling yields at pH 2 and at pH >6. The radiolabeling kinetics of each of the BFCs was examined with 67Ga and 68Ga. Table 1 summarizes the optimal reaction conditions, with respect to radiochemical yield and specific activity, determined for each BFC. The BFC concentration had
the largest impact on reaction time and radiochemical yield. At higher concentrations (>30 µM), all the BFCs achieve >95% radiochemical yield after only 5 min at room temperature. At lower concentrations, slightly longer reaction times (10 min) are required to obtain >95% radiochemical yield for 67Ga-pNO2-Bn-PCTA and p-NO2-Bn-Oxo, but p-NO2-Bn-DOTA required significantly longer reaction times (60 min). A similar trend was observed for 68Ga radiolabeling. p-NO2-Bn-PCTA, p-NO2-Bn-Oxo, and p-NO2-Bn-NOTA were efficiently labeled at room temperature, even when using low concentrations of BFC, while p-NO2-Bn-DOTA had slower reaction kinetics. Heating the reaction at 80 °C was necessary to obtain 68Ga-pNO2-Bn-DOTA in radiochemical yields and specific activities comparable to those obtained for the other BFCs at room temperature. Stability. The stability of the Ga-radiolabeled BFCs was assessed in different buffer solutions and in the presence of excess apo-transferrin. No degradation or loss of Ga was observed in the buffered solutions at pH 4 and higher. After 1 h incubating at pH 2, the 67Ga radiolabled BFCs showed negligible degradation or loss of 67Ga, but after 24 h, 67Ga radiolabeled p-NO2-Bn-Oxo and p-NO2-Bn-DOTA showed a 20% and 36% loss, respectively, of 67Ga from the BFC. The kinetic stability in the presence of apo-transferrin for each of the BFCs followed the same trend, whether radiolabeled with 67 Ga or 68Ga. p-NO2-Bn-NOTA allowed negligible transfer of the radioisotope to the protein. p-NO2-Bn-DOTA, and to a lesser extent p-NO2-Bn-PCTA, allowed some loss of radioisotope to apo-transferrin over the 4 h time period, while p-NO2-Bn-Oxo was kinetically labile, with almost all of the radioisotope transferred to the protein in less than 15 min. Biodistribution. The %ID/g for select tissues was determined for each of the 68Ga radiolabeled BFCs in mice at 15 min, 1 h, and 4 h post injection; the results are summarized in Figures 4-8. The majority of activity was found in the kidneys and urine, indicating renal clearance. 68Ga-radiolabeled p-NO2-BnNOTA had significantly more activity in the kidney at all time points, 4-14 times greater than the kidney %ID/g of 68Garadiolabeled p-NO2-Bn-PCTA or p-NO2-Bn-DOTA. The %ID/g of all tissues for 68Ga-radiolabeled p-NO2-Bn-PCTA and p-NO2Bn-DOTA were not statistically different. For the liver, intestine, and brain tissue, 68Ga-radiolabeled p-NO2-Bn-NOTA, p-NO2Bn-PCTA, and p-NO2-Bn-DOTA were similar, but for blood, heart, muscle, bone, and spleen tissue 68Ga p-NO2-Bn-NOTA cleared more rapidly with statistically (p < 0.05) lower activity after 1 h; 80-90% of the activity at 15 min was cleared by 1 h for 68Ga-radiolabeled p-NO2-Bn-NOTA compared to 45-75% of the activity for 68Ga-radiolableled p-NO2-Bn-PCTA and p-NO2-Bn-DOTA in these tissues. In contrast to the other BFCs, 68 Ga-radiolabeled p-NO2-Bn-Oxo did not appreciably clear from
534 Bioconjugate Chem., Vol. 21, No. 3, 2010
Ferreira et al.
Table 1. Optimized Reaction Conditions for 67Ga and 68Ga Radiolabeling of BFCsa bifunctional chelate p-NO2-Bn-PCTA
isotope 67
Ga Ga Ga 67 Ga 67 Ga 67 Ga 68 Ga 68 Ga 68 Ga 68 Ga 68 Ga 68 Ga 67
p-NO2-Bn-Oxo p-NO2-Bn-NOTA p-NO2-Bn-DOTA p-NO2-Bn-PCTA p-NO2-Bn-Oxo p-NO2-Bn-NOTA p-NO2-Bn-DOTA
67
reaction conditions ([BFC], temperature, time)
radiochemical yield
specific activity (mCi/µmol)
3 µM, RT, 10 min 15 µM, RT, 5 min 3 µM, RT, 10 min 3 µM, RT, 5 min 3 µM, RT, 60 min 20 µM, RT, 30 min 1 µM, RT, 5 min 1 µM, RT, 5 min 1 µM, RT, 5 min 1 µM, 80 °C, 10 min 10 µM, RT, 30 min 10 µM, 80 °C, 5 min
96.9% 98.4% 98.4% 99.9% 97.4% 96.4% 98.9% 98.5% 98.6% 92.5% 87.5% 95.2%
166 35 163 167 162 25 499 496 488 472 45 48
a All reactions done in sodium acetate buffer pH 4-5, RT ) room temperature, specific activity without purification reported as mCi of isotope incorporated/µmol of BFC.
Figure 4. 68Ga-p-NO2-Bn-Oxo biodistribution in CD1 mice over 4 h.
Figure 5. 68Ga-p-NO2-Bn-PCTA biodistribution in male CD1 mice (n ) 4) over 4 h.
tissue, other than the kidneys, and in some tissues, such as bone, muscle, and intestines, the %ID/g increased over time. PET imaging experiments were in agreement with the biodistribution studies. The kidney and bladder were clearly visible and contained the highest activity for all of the 68Garadiolabeled BFCs. The activity per unit volume measured in the kidney was on average 4.5 times higher for the 68Garadiolabeled p-NO2-Bn-NOTA compared to the other 68Garadiolabeled BFCs. While no other tissues could be delineated for 68Ga-radiolabeled p-NO2-Bn-NOTA, heart and intestine were visible for the other three 68Ga-radiolabeled BFCs at early time points and were most notable in the images of mice injected with 68Ga-radiolabeled p-NO2-Bn-Oxo.
DISCUSSION To determine if the novel BFCs, p-NO2-Bn-PCTA and p-NO2-Bn-Oxo, offered any advantage for Ga-based radio-
Figure 6. 68Ga-p-NO2-Bn-DOTA biodistribution in male CD1 mice (n ) 4) over 4 h.
Figure 7. 68Ga-p-NO2-Bn-NOTA biodistribution in male CD1 mice (n ) 4) over 4 h.
pharmaceutical development, they were directly compared to the currently used BFCs, p-NO2-Bn-PCTA DOTA and p-NO2Bn-NOTA. Numerous other reports of radiolabeling and animal studies with DOTA and NOTA containing compounds typically focus on targeted conjugates and not the BFC alone. As well, different isotope suppliers, methods, and reagents can all impact experimental results. So, a direct comparison of the novel BFCs to the analogous DOTA and NOTA containing BFCs was warranted to best evaluate p-NO2-Bn-PCTA and p-NO2-Bn-Oxo for use with Ga radioisotopes. Efficient incorporation of the radioisotope in high yields and high specific activity is particularly important for 68Ga, due to the short half-life of radioisotope. If the radiolabeling reaction were to yield product with high radiochemical purity and specific activity, post-radiolabeling purification could be obviated. The need for purification of 68Ga-radiolabeled peptides (9, 11-13, 15),
Evaluation of Bifunctional Chelates for Gallium
Figure 8. 68Ga-BFC %ID/g in urine in male CD1 mice (n ) 4) over 4 h.
typically done by HPLC, is not optimal, as it is time-consuming and not necessarily accessible in clinical settings. As some targeting vectors may be sensitive to heat, organic solvents, or extreme pH, radiolabeling reactions that can be done under mild conditions, such as in near neutral aqueous buffers and at room temperature, are preferred. Comparison of the four BFCs suggested that for labeling with gallium radioisotopes the two novel BFCs, p-NO2-Bn-PCTA and p-NO2-Bn-Oxo, as well as p-NO2-Bn-NOTA, were all superior to p-NO2-Bn-DOTA. The radiolabeling efficiency of p-NO2-Bn-DOTA was greatly impacted by pH, required longer reaction times or heating, and gave lower yields and lower specific activities. Radiolabeling of DOTA conjugates has typically been done at elevated temperatures (7, 9, 10), and the slow radioisotope incorporation at room temperature has also been reported (8). In contrast, NOTA and NOTA conjugates have recently been reported to radiolabel efficiently under ambient conditions (14). The two novel BFCs, p-NO2-Bn-PCTA and p-NO2-Bn-Oxo, also shared this property. Using either 67Ga or 68Ga radiolabeling at room temperature, p-NO2-Bn-PCTA, p-NO2-Bn-Oxo, and p-NO2-Bn-NOTA were all efficiently radiolabeled with 67Ga or 68Ga in minutes without the need for heating or purification. Similar trends were observed whether the radiochemistry was done with 67Ga or 68Ga, as would be expected, but the results were not identical. Differences observed in radiolabeling may be due to differences in the level of trace metal impurities found in the isotope. p-NO2-Bn-DOTA was found to have slower reaction kinetics for 68Ga labeling compared to 67Ga labeling. Trace metals that are coeluted from the 68Ge generator can compete with 68Ga for the BFC (7, 17). It has been suggested that heating may improve the selectivity of the BFCs for 68Ga over other trace metals (17), which is supported by the high radiochemical yields and specific activities obtained when the 68 Ga radiolabeling of p-NO2-Bn-DOTA was performed at 80 °C. The 68Ga radiolabeling efficiency of p-NO2-Bn-NOTA was not impacted by the increased level of metal impurities, possibly due to the smaller macrocycle having a greater preference for the small Ga3+ ion. The greater stability of p-NO2-Bn-PCTA and p-NO2-BnNOTA was illustrated in the transferrin challenge study. Transferrin is an iron transfer protein found in blood. Due to the similar ionic radius and charge of Fe3+ (0.645 Å) and Ga3+ (0.620 Å) (18), 67Ga or 68Ga are known to bind to transferrin (1). In the transferrin challenge experiment, each of the Garadiolabeled BFCs was incubated in the presence of excess apotransferrin, and the amount of Ga lost from the BFC and bound to transferrin was monitored over time. Both Ga-radiolabeled p-NO2-Bn-NOTA and p-NO2-Bn-PCTA proved to be relatively inert over 4 h, which is an appropriate period for imaging with 68 Ga. Ga-radiolabeled p-NO2-Bn-DOTA showed some loss of
Bioconjugate Chem., Vol. 21, No. 3, 2010 535
Ga to apo-transferrin over time. The instability of Ga-radiolabeled p-NO2-Bn-DOTA in the presence of serum proteins, such as transferrin, has been previously reported (8, 9). Ga-radiolabeled p-NO2-Bn-Oxo was determined to be kinetically labile and was predicted to be unstable in vivo, despite having a thermodynamic stability constant similar to p-NO2-Bn-DOTA (Ga-p-NO2-Bn-Oxo pK ) 21.3 (19), Ga-p-NO2-Bn-DOTA pK ) 21.33 (20)). The biodistribution results for Ga-radiolabeled p-NO2-Bn-Oxo supported the in ViVo instability of this species, as negligible clearance from the blood and other tissues was observed, as well as 68Ga accumulation in bone and muscle tissue. Mouse biodistribution and imaging studies with the three stable 68Ga-radiolabeled BFCs, p-NO2-Bn-PCTA, p-NO2-BnNOTA, and p-NO2-Bn-DOTA, provided information on the effect the BFC has on background uptake and clearance. Because of the short half-life of 68Ga, imaging is done soon after injection leaving minimal time for the radiopharmaceutical to clear from blood and other tissue. All three of the BFCs cleared over time, mainly through the kidneys. For all tissues, the %ID/g values for 68Ga-radiolabeled p-NO2-Bn-PCTA compared to 68Ga-radiolabeled p-NO2-Bn-DOTA were not statistically different. The biodistributions of these two BFCs were similar despite the structural differences, the slight differences noted in stability, and the difference in charge (at physiological pH, Ga-p-NO2-Bn-PCTA is expected to be neutral, while Gap-NO2-Bn-DOTA would be monoanionic) (21, 22). While the biodistribution of 68Ga-radiolabeld p-NO2-Bn-NOTA was statistically similar to those of 68Ga-radiolabled p-NO2-Bn-PCTA and p-NO2-Bn-DOTA with respect to uptake and clearance in the intestine, liver, and bone, large differences were observed for all other tissues. 68Ga-radiolabeled p-NO2-Bn-NOTA cleared much more rapidly from the blood, heart, muscle, and spleen, but had 5 to 20 times more activity retained in the kidney and bladder (urine). The fast clearance from ubiquitous background tissue, such as blood and muscle, appears advantageous for obtaining superior target to background ratios, as long as the imaging target is not in proximity of the kidneys or bladder. Contrastingly, the fast blood clearance may limit the accumulation of the Ga-radiolabeled agent by limiting the circulation time of the agent. By 15 min, 68Ga-radiolabeled p-NO2-Bn-NOTA in the blood (%ID/g) is already half that of 68Ga-radiolabeled p-NO2-Bn-PCTA and p-NO2-Bn-DOTA, and by 1 h, 68Garadiolabeled p-NO2-Bn-NOTA has almost completely cleared from the blood. 68Ga-radiolabeled p-NO2-Bn-PCTA and p-NO2Bn-DOTA also clear from the blood and related tissues, but at a slower rate. The need for relatively fast blood clearance may need to be balanced with adequate circulation time to ensure target accumulation. The addition of a targeting vector such as a peptide is expected to impact the nontargeted tissue distribution. Dependent on the nature and size of the targeting vector, the blood circulation time of the agent may increase and more activity may be retained in the kidney and/or liver due to metabolic processes. Further studies comparing targeted-peptideconjugated BFCs for molecular imaging are planned to determine how differences in the clearance rates of blood and other tissues affect target uptake and target to background ratios.
CONCLUSIONS In this study, we evaluated two novel BFCs, p-NO2-Bn-PCTA and p-NO2-Bn-Oxo, for Ga-based radiopharmaceutical development by direct comparison to the commonly used BFCs, p-NO2Bn-DOTA and p-NO2-Bn-NOTA. Of the four BFCs, p-NO2Bn-NOTA and p-NO2-Bn-PCTA were found to be more efficiently radiolabeled under milder conditions and to yield kinetically inert complexes that have appropriate clearance properties in vivo. The biodistribution of 68Ga-radiolabeled
536 Bioconjugate Chem., Vol. 21, No. 3, 2010
p-NO2-Bn-PCTA was not statistically different than that of 68Garadiolabeled p-NO2-Bn-DOTA. However, p-NO2-Bn-PCTA could be radiolabeled with Ga more efficiently under mild conditions compared to p-NO2-Bn-DOTA. Due to superior Garadiolabeling efficiency, p-NO2-Bn-PCTA may be an improved alternative to p-NO2-Bn-DOTA in Ga isotope based radiopharmaceutical development. Radiolabeling with p-NO2-Bn-NOTA gave high radiochemical yields in short reactions times and could be done at room temperature. Fast clearance from most tissue was observed for 68Ga-p-NO2-Bn-NOTA, but considerably higher retention in the kidneys was also noted. While the fast background clearance is advantageous, the fast blood clearance of 68Ga-p-NO2-Bn-NOTA could potentially limit target accumulation and requires further investigation.
LITERATURE CITED (1) Weiner, R. E. (1996) The mechanism of 67Ga localization in malignant disease. Nucl. Med. Biol. 23, 745–752. (2) Weiner, R. E., and Thakur, M. L. (2003) Chemistry of galllium and indium radiopharmaceuticals, in Handbook of Radiopharmaceuticas Radiochemistry and Applications (Welch, M., and Redvanly, C. S., Eds.), John Wiley and Sons Ltd., West Sussex, England. (3) Khan, M. U., Khan, S., El-Refaie, S., Win, Z., Rubello, D., and Al-Nahhas, A. (2009) Clinical indication for Gallium-68 positron emission tomography imaging. EJSO 35, 561–567. (4) Fani, M., Andre, J. P., and Maecke, H. (2008) 68Ga-PET: a powerful generator-based alternative to cyclotron-based PET radiopharmacueticals. Contr. Med. Mol. Imaging 3, 53–62. (5) Maecke, H., Hofmann, M., and Haberkorn, U. (2005) 68Galabeled peptides in tumor imaging. J. Nucl. Med. 46, 172S–178S. (6) Fichna, J., and Janecka, A. (2003) Synthesis of target-specific radiolabeled peptides in diagnostic imaging. Bioconjugate Chem. 14, 3–17. (7) Velikyan, I., Beyer, G. J., and Langstrom, B. (2004) Microwavesupported preparation of 68Ga bioconjugates with high sepcific radioactivity. Bioconjugate Chem. 15, 554–560. (8) Blom, E., Langstrom, B., and Velikyan, I. (2009) 68Ga-Labeling of biotin analogues and their characterization. Bioconjugate Chem. 20, 1146–1151. (9) Decristoforo, C., Gonzalez, I. H., Carlsen, J., Rupprich, M., Huisman, M., Virgolini, I., Wester, H.-J., and Haubner, R. (2008) 68Ga- and 111Inlabelled DOTA-RGD peptides for imaging Rvβ3 integrin expression. Eur. J. Nucl. Med. Mol. Imaging 35, 1507–1515. (10) Griffiths, G. L., Chang, C.-H., McBride, W. J., Rossi, E. A., Sheerin, A., Tejada, G. R., Karacay, H., Sharkey, R. M., Horak, I. D., Hansen, H. J., and Goldenberg, D. M. (2004) Reagents and methods for PET using bispecific antibody pretargeting and 68 Ga-radiolabeled bivalent hapten-peptide-chelate conjugates. J. Nucl. Med. 45, 30–39.
Ferreira et al. (11) Ren, G., Zhang, R., Liu, Z., Webster, J. M., Miao, Z., Gambhir, S. S., Syud, F. A., and Cheng, Z. (2009) A 2-helix small protein labeled with 68Ga for PET imaging of HER2 expression. J. Nucl. Med. 50, 1492–1499. (12) Liu, Z., Yan, Y., Liu, S., Wang, F., and Chen, X. (2009) 18F, 64 Cu, and 68Ga labeled RGD-bombesin heterodimeric peptides for PET imaging of breast cancer. Bioconjugate Chem. 20, 1016– 1025. (13) Li, Z.-B., Chen, K., and Chen, X. (2008) 68Ga-labeled multimeric RGD peptides for microPET imaging of integrin Rvβ3 expression. Eur. J. Nucl. Med. Mol. Imaging 35, 1100– 1108. (14) Velikyan, I., Maecke, H., and Langstrom, B. (2008) Convenient preparation of Ga-68 based PET-radiopharmaceuticals at room temperature. Bioconjugate Chem. 19, 569–573. (15) Jeong, J. M., Hong, M. K., Chang, Y. S., Lee, Y.-S., Kim, Y.-J., Cheon, G. J., Lee, D. S., Chung, J.-K., and Lee, M. C. (2008) Preparation of a promising angiogenesis PET imaging agent: 68Ga-Labeled c(RGDyK)-Isothiocyanatobenzyl-1,4,7-triazacyclononane-1,4,7-triacetic acid and feasibility studies in mice. J. Nucl. Med. 49, 830–836. (16) Ferreira, C. L., Yapp, D. T., Lamsa, E., Gleave, M., Bensimon, C., Jurek, P., and Kiefer, G. E. (2008) Evaluation of novel bifunctional chelates for development of Cu-64 based radiopharmaceuticals. Nucl. Med. Biol. 35, 875–882. (17) Velikyan, I., Beyer, G. J., Bergstrom-Pettermann, E., Johansen, P., Bergstrom, M., and Langstrom, B. (2008) The importance of high specific radioactivity in the performance of 68Ga-labeled peptide. Nucl. Med. Biol. 35, 529–536. (18) Shannon, R. D. (1976) Revised effective radii and systematic studies of interatomic distances in halide and chalcogenides. Acta Crystallogr. A32, 751–767. (19) Delgado, R., Quintino, S., Teixeira, M., and Zhang, A. (1996) Metal complexes of a 12-membered tetraaza macrocycle containing pyridine and N-carboxymethyl groups. J. Chem. Soc., Dalton Trans. 55–63. (20) Clarke, E. T., and Martell, A. E. (1991) Stabilities of trivalent metal ion complexes of the tetraacetate derivatives of 12-, 13and 14-membered tetraazamacrocycles. Inorg. Chim. Acta 190, 37–46. (21) Tircso, G., Benyo, E. T., Suh, E. H., Jurek, P., Kiefer, G. E., Sherry, A. D., and Kovacs, Z. (2009) (S)-5-(p-Nitrobenzyl)PCTA, a promising bifunctional ligand with advantageous metal ion complexation kinetics. Bioconjugate Chem. 20, 565– 575. (22) Woods, M., Kovacs, Z., Kiraly, R., Brucher, E., Zhanng, S. R., and Sherry, A. D. (2004) Solution dynamics and stability of lanthanide(III) (S)-2-(p-nitrobenzyl)DOTA complexes. Inorg. Chem. 43, 2845–2851. BC900443A
