Abstract
HER2/neu and MUC1-based synthetic peptides were prepared and evaluated in an effort to develop peptide-based radiopharmaceuticals derived from tumor-associated-antigens for the detection of breast cancer. The receptors for HER2/neu and MUC1 are overexpressed in various human cancers, such as breast and ovarian cancer. The relatively low expression of these antigens on normal tissues makes them attractive targets for tumor imaging. In addition, antitumor-antibody-derived peptides based on the Glu-Pro-Pro-Thr (EPPT) sequence were prepared for the detection of breast cancer. It has been shown that the EPPT peptide has high affinity for the MUC1-derived peptide. The peptides were prepared by solid-phase synthesis and radiolabeled efficiently with 99mTc via ligand exchange. They exhibited good stability in vitro in human plasma and against cysteine and histidine challenge. The peptides displayed high affinities (in nanomolar range) for MCF-7, MDA-MB-231 and T47-D breast cancer cell lines in vitro. Additionally, they exhibited a rapid internalization into tumor cells. In vivo biodistribution in mice showed rapid and efficient blood clearance and excretion mainly through the renal-urinary route, with some elimination via the hepatobiliary pathway. However, the extent of urinary excretion was found to be variable for radiopeptides, with the highest being observed for antitumor-antibody-derived peptide. The peptides showed moderate tumor uptake (up to 2.2±0.98% ID/g) in nude mice carrying breast tumor xenografts. The uptake in the tumor was always higher than in the blood and muscle. A fast clearance from the blood and low accumulation (<6% ID/g) by the major organs was obtained in nude mice resulting in favorable tumor/blood and tumor/muscle ratios as early as 1 h after injection. The combination of favorable in vitro and in vivo characteristics makes this new and interesting class of peptides potential candidates for the diagnosis of breast cancer in vivo.
Many tumor-associated antigens (TAAs) have been discovered and identified in the last decade and have provided new hope for the treatment of patients with malignant disease (1, 2). For the potential treatment of breast and ovarian cancer, however, only a few T-cell epitopes have been identified, including those derived from the HER2/neu proto-oncogene and the epithelial mucin, MUC1 (1-3). Most tumor antigens are not only expressed in cancer cells but also in normal cells. One problem with this is that if tumor immunity is focused on antigens which are also expressed in normal cells (i.e. autoantigens), then any effect on tumor cells may cause an autoimmune response (4). Some tumor-specific antigens are derived from proteins that are structurally abnormal or completely silent in normal tissues. The human epithelial mucin encoded by the gene MUC1 is an example of a tumor-specific antigen that is silent on normal tissues but it is overexpressed on almost all human epithelial cell cancers, including >90% of human breast, ovarian, pancreatic, colorectal, lung, prostate and gastric cancers (2, 5), thus making MUC1 a promising tumor-antigen with diagnostic as well as therapeutic potential in the management and treatment of cancer (5, 6). Another TAA that is amplified and/or overexpressed in several human cancers including breast and ovarian cancer is HER2 (human epidermal growth factor receptor type 2) also called HER2/neu, the gene product of c-erbB-2 proto-oncogene (1, 3). Gene amplification and overexpression of HER2/neu have been demonstrated with high frequency in a number of human malignancies including breast and ovarian tumors (7). The overexpression of HER2/neu in various tumor cells and relatively low expression of this antigen in normal tissues (8-11), may make it an attractive target for tumor imaging with suitably designed HER2/neu-derived synthetic peptides. The frequency of HER2/neu overexpression in various cancers and its importance in mediating cancer cell growth and aggressiveness have resulted in the development of a humanized monoclonal antibody, trastuzumab (Herceptin™). Blocking of HER2/neu signaling in breast cancers using the anti-HER2/neu antibody trastuzumab results in growth inhibition of tumor cells and an increase in the patients' survival rate (7, 11). Herceptin™ has been approved for clinical use in the U.S.A. and in Europe for the treatment of metastatic breast cancer (11, 12).
Scintigraphic detection of HER2/neu-positive breast cancer may provide important diagnostic information influencing patient management. The use of radionuclide diagnostic imaging would enable detection of HER2/neu expression by a non-invasive procedure in both primary tumors and metastases, without false-negative results due to biopsy sampling errors (11). Clinical utility of radionuclide imaging of HER2/neu expression in breast cancer patients has been demonstrated recently by using 111In-DTPA-trastuzumab (13-15). Moreover, radionuclide targeting of HER2/neu, for both imaging and therapy, using intact antibodies has already been evaluated (11, 15-20). However, slow tumor targeting and slow blood clearance of intact antibodies are the well-known problems, and the use of small targeting agents seems to be crucial in future nuclear medicine, at least for diagnostic imaging (11, 21). The use of synthetic peptides in cancer diagnosis offers practical advantages over antibodies, such as relative ease of construction and production, chemical stability and a lack of infectious or oncogenic potential (10, 22, 23). It was assumed that synthetic peptides derived from TAAs and retaining most of the original affinity and specificity to tumor-antigens would be more suitable for tumor imaging. Though no information is available about the presence of specific receptor and/or binding affinity for the tumor-antigen-derived peptides HER2/neu and MUC1, there was an interest to determine whether these tumor-antigen-derived peptides have the ability to bind to breast cancer cells that express receptors for these peptides. In an effort to develop peptides derived from TAAs for imaging breast cancer, novel HER2/neu and MUC1-derived peptides were synthesized together with antitumor-antibody derived EPPT peptides and were coupled with MAG3 (mercaptoacetyltriglycine) chelator for radiolabeling with technetium-99m (99mTc). 99mTc is the radionuclide of choice in the development of imaging agents by virtue of its wide availability (from 99mTc-generator system), short half-life (6 h), and ideal energy (140 κeV) for γ-imaging (23). MAG3 chelating moiety is used because of its feasibility of coupling to the targeting peptide molecules during peptide synthesis and its well-defined chemistry (24). The synthesized peptides include: HER2/neu-derived peptide (MAG3-HER2/neu, MAG3-(D)Asp1-Lys2-Ile3-(D)Phe4-Gly5-Ser6-Leu7-Ala8-(D)Phe9-Leu10-CONH2) (3, 25-27), and the peptide derived from the breast cancer-associated mucin, MUC1 (MAG3-MUC1, MAG3-(D)Asp1-Pro2-(D)Asp3-Thr4-Arg5-Pro6-CONH2) (5, 28). Moreover, a hybrid peptide, which combines the important amino acid sequences of two distinct TAAs, HER2/neu and MUC1 (MAG3-HER2/neu-MUC1, MAG3-(D)Asp1-Lys2-Ile3-(D)Phe4-Gly5-Ser6-Leu7-Ala8-(D)Phe9-Leu10-Pro11-(D)Asp12-Thr13-Arg14-Pro15-CONH2) was synthesized. In addition, a small peptide was prepared (MAG3-EPPT, MAG3-(D)Asp1-Glu2-Pro3-Pro4-Thr5-CONH2), derived from the CDR3 VH region of a tumor-associated monoclonal antibody (ASM2), raised against human epithelial cancer cells. The immunological affinity of this peptide is mainly based on the EPPT residues in its sequence (5, 29, 30). It has been shown that the synthetic EPPT peptide has significant affinity for the MUC1-derived peptides (5). A hybrid peptide containing the critical amino acid sequences of both EPPT and MUC1 was also synthesized (MAG3-EPPT-MUC1, MAG3-(D)Asp1-Glu2-Pro3-Pro4-Thr5-Pro6-(D)Asp7-Thr8-Arg9-Pro10-CONH2), and its potential as a tumor imaging agent was explored. In a previous study, the full length EPPT peptide was evaluated for the detection of breast cancer in vivo (29).
Herein, the continued research on the development of new peptide-based radiopharmaceuticals for tumor imaging, the synthesis, radiolabeling with 99mTc, and in vitro and in vivo characterization of a new and interesting class of radiopharmaceuticals derived from the TAAs and antitumor-antibody for the detection of breast cancer are described.
Materials and Methods
Preparation of peptides. All peptides were prepared by solid-phase synthetic techniques on a CS036 peptide synthesizer (CS Bio Co., San Carlos, CA, USA), following standard Fmoc (9-fluorenylmethoxycarbonyl) chemistry (31), using Rink amide MBHA (4-methylbenzhydrylamine) resin according to a general method described previously (29). Briefly, after stepwise incorporation of all the desired amino acids to the sequence, the last N-terminal Fmoc-protecting group was removed and S-benzoylmercaptoacetic acid (synthesized by reacting mercaptoacetic acid with benzoic acid chloride in the presence of a base) (32), was manually coupled to the free N-terminus of each peptide to yield MAG3-peptide conjugates. The purity of each peptide was confirmed by HPLC and their identity by mass spectrometry.
Radiolabeling with 99mTc. A volume of 25-50 μL of each peptide (1 mg/mL CH3CN/H2O solution) was mixed with 200 μL 0.5 M citrate-phosphate buffer (pH 9-11), 250 μL (10 mg) sodium potassium tartrate solution and 20 μL 5% ascorbic acid. To this, 100 μL (100 μg) of SnCl2.2H2O in 0.05 N HCl were added followed by 100-300 μL of 99mTcO4- (5-15 mCi). The labeling mixture was then heated at 90°C for 10 min. The final pH of the reaction mixtures was between 7 and 9.
HPLC purification and analysis. HPLC analysis and purification of all five peptides and their metal complexes were performed on a Shimadzu HPLC system (Shimadzu Corporation, Kyoto, Japan) using a C18 reversed-phase column (5 μm, 3.9×150 mm) (DeltaPak, Waters Corporation, Milford, MA, USA). HPLC was run using a gradient system of 0.1% (v/v) TFA in water and 0.1% (v/v) TFA in acetonitrile at a flow rate of 0.9 mL/min. For each peptide, the major peak was isolated and the solvent was evaporated. The HPLC-purified compounds were reconstituted in sterile saline and used for cell-binding and animal biodistribution studies.
Cysteine and histidine challenge. The bond strength of 99mTc-complexes was determined by displacement of peptide bound radioactivity with cysteine and histidine. A known amount of each radiopeptide was incubated with an excess amount of cysteine or histidine (100-fold molar excess compared to the peptide) at 37°C for up to 4 h. For transchelation studies, peptides were not purified by HPLC after labeling, hence, some tin, 99mTcO4- or unlabeled peptides remained present in the solutions. The samples were analyzed at 1 and 4 h after incubation by HPLC and the percentage radioactivity associated with the cysteine or histidine and the intact radiolabeled peptides were determined.
Octanol-saline partition coefficient. To determine the lipophilicity, HPLC-purified radiopeptides (100 μL, ~100 μCi) were added to separate test tubes each containing 1 mL of n-octanol and 1 mL of saline. The tubes were vortexed for 1 min. After partial separation of the phases by gravity for another 15 min, 0.5 mL of each phase was transferred to separate tubes and centrifuged. Duplicate 100 μL aliquots of each phase were measured and the partition coefficient was determined by the function:
Partition coefficient=Log10 (radioactivity in octanol layer/radioactivity in aqueous layer) (33).
In vitro stability in plasma and in T47-D cells. Each radiopeptide (100 μL, ~500 μCi) was incubated with human plasma (500 μL) in duplicate at 37°C for up to 4 h. Following incubation at 1 and 4 h, the plasma proteins were precipitated with a mixture of acetonitrile/ethanol (1:1 v/v) and the sample was centrifuged. The supernatant layer was filtered through a 0.2-μm pore filter and analyzed by HPLC. The stability in the cell culture was examined by incubating each 99mTc-peptide (50 μL, 100 μCi) with T47-D cells suspension (~200,000 cells in 200 μL) in phosphate-buffered saline (PBS) at 37°C for 1 h. The mixtures were then centrifuged and the supernatant layers were analyzed by HPLC.
In vitro cell-binding and internalization. The cell-binding and subsequent internalization into MCF-7, MDA-MB-231 and T47-D breast cancer cell lines (American Type Culture Collection, Rockville, MD, USA) of peptides were performed according to the method described previously with a few modifications (34). In brief, approximately 300,000 cells (in 300 μL PBS) were incubated with various amounts of peptides (prepared from the serial dilutions of HPLC-purified radiopeptides) in duplicate for 1 h at room temperature. Incubation was terminated by dilution with cold saline (300 μL) and the cells were pelleted by centrifugation. The cell-pellets were rapidly washed with cold saline to remove any unbound peptide and centrifuged to collect supernatants. Radioactivity in the cell-pellet and the washings was measured in a γ-counter. Non-specific binding was determined in the presence of approximately 100-fold excess of unlabeled peptide. Specific binding was calculated by subtracting the non-specifically bound radioactivity from that of the total binding. The data were analyzed (to estimate respective binding affinities) by a non-linear regression analysis program (Phillip H. Sherrod, Brentwood, TN, USA) using one-site binding equation (35). Thereafter, the cell pellets were incubated again with 300 μL of acidic buffer (0.02 M sodium acetate in saline, pH 5.0) (36) at 37°C for 20 min to allow cellular internalization of the surface bound radioligand (34, 37). After incubation, cells were separated by centrifugation. The amount of cell surface-bound (acid-wash) and intracellular radioactivity (acid-resistant) was determined by measuring the radioactivity of the supernatants and the cell-pellets, respectively, in a γ-counter (34).
In vivo animal biodistribution. Prior approval was obtained from the institutional Animal Safety and Care Committee for the use of animals. Animal studies were conducted according to the international regulations governing the safe and proper use of laboratory animals (38). In vivo biodistribution studies were performed on healthy Balb/c mice (body mass 18-25 g) and nude mice implanted with MDA-MB-231 cells. The HPLC-purified radioligands (100 μL, ~10 μCi, total peptide dose ≤1 μg) were injected intravenously via the lateral tail vein and the animals (n=3 animals per time point) were sacrificed by neck dislocation at 1 and 4 h post-injection (p.i.). The blood and urine were immediately collected. Any urine expelled on sacrificing the animals was measured together with that remaining in the bladder. Major organs (lungs, stomach, liver, kidneys and intestines) were excised, weighed and their radioactivity measured in a γ-counter. The percentage injected dose per gram (% ID/g) of tissue was calculated by counting the samples in a γ-counter against the suitably diluted aliquots of the injected solution as a standard. For clearance studies, radioactivity in the urine with bladder and intestines with contents is reported as the percentage of the injected dose (% ID). For the induction of tumor xenografts, estrogen-independent and highly aggressive MDA-MB-231 cells were subcutaneously injected at a concentration of 1×107 cells per mouse and allowed to grow in vivo for 3-4 weeks. After adequate growth of the tumors (0.6-1.0 g), tumor uptake studies were performed and the amount of radioactivity in the tumor and other major organs was determined by γ-counting as stated above for Balb/c mice.
Results and Discussion
Tumor targeting with small radiolabeled receptor-binding peptides is an attractive approach for the diagnosis and therapy of cancer (23). Many human tumors overexpress a variety of receptors for various peptides, which is the basis of peptide receptor-targeted imaging in nuclear medicine. The rapid pharmacokinetics and high affinity of radiopeptides for receptors on tumor cells can be effectively used for diagnosis and peptidic radionuclide therapy (23). Receptors for tumor-antigens, such as HER2/neu and MUC1, are commonly found preferentially on various tumor cells than in normal cells (4, 10, 11, 16), providing the basis for successful use of tumor-antigen-derived peptides for tumor imaging. One attractive way for developing compounds for targeting HER2/neu- and MUC1-positive tumors is the use of peptides derived from these tumor-specific antigens. An advantage of such peptide molecules is that they can be easily synthesized chemically and modified molecularly according to pharmacokinetic requirements. In this study, biologically active minimal sequences of human-tumor-derived peptides were designed for easy preparation and handling in order to evaluate their potential as breast cancer imaging agents.
Preparation of peptides. All peptides investigated in this study were conveniently and successfully prepared by solid-phase peptide synthesis using standard Fmoc/HBTU chemistry (31). It has been shown that the structure and composition of the chelating moiety and the spacer group linking the chelator to the peptide play a vital role in determining the physiochemical and pharmacokinetic characteristics of the resultant bioconjugate (39). A well-established renal imaging agent, MAG3, was used as a chelator, and a hydrophilic aspartic acid residue was used as a spacer, in an effort to enhance urinary excretion and minimize hepatobiliary elimination. MAG3 is introduced at the N-terminus by manual synthesis. The identity and purity of each peptide was confirmed by mass spectrometry and HPLC. It was found that the experimentally determined molecular weights correlated well with the theoretically calculated values (Table I). The synthetic yields were determined to be 20-30% by RP-HPLC analysis.
Radiolabeling with 99mTc. Radiolabeling of peptides with 99mTc was achieved by the standard ligand exchange method using sodium potassium tartrate as a weak complexing agent. The radio-HPLC profiles of all radiopeptides showed the formation of one main radioactive compound with some minor peaks. An example of radiolabeling MAG3-EPPT-MUC1 with 99mTc is presented in Figure 1. By the exchange labeling approach, the radiolabeling efficiency of 99mTc-labeled peptides ranged between 65 and 98% (Table II), with a specific radioactivity of greater than 100 Ci/mmol. Usually, the amounts of 99mTc-peptides produced were below the UV-detection limit during HPLC analysis, which somewhat impeded the exact estimation of the specific radioactivity of the respective radioligands. The extent of 99mTc-complexation and the radiolabeling yield were found to be pH dependent and the optimal radiolabeling of all the peptides was achieved in the presence of 0.5 M citrate-phosphate buffer pH 10 or higher. This is expected because an alkaline pH facilitates the deprotonation of amide-nitrogens in MAG3 and the removal of the thiol-protecting group. Both these effects are known to enhance complexation of MAG3 with 99m Tc (24). The stability of the 99mTc-conjugates was determined by HPLC at different time intervals. The radiopeptides remained stable (up to 70%) at room temperature for up to 18 h. The HPLC retention times, ranging from 12 to 19.3 min, reflected the expected lipophilicity of the 99mTc-complexes. 99mTc-MAG3-EPPT, the smallest peptide studied, displayed the shortest retention time (12.0 min), hence the lowest lipophilicity (log P=-3.15±0.05) (Table II), whereas 99mTc-MAG3-HER2/neu, with the highest retention time of 19.3 min, showed the highest lipophilicity (log P=1.31±0.11) among the tested compounds.
Cysteine and histidine challenge. All radiopeptides were challenged with excess cysteine and histidine in order to determine their stability against ligand exchange and/or decomposition. A known amount of each peptide was incubated with excess of cysteine or histidine and the samples were analyzed by HPLC at 1 and 4 h. The results are summarized in Table III. A significant reoxidation (up to 30%) of the 99mTc-complexes to 99mTcO4- was detected even after 1 h at 37°C, indicating a somewhat weak bond strength of the 99mTc-complexes. In contrast, these complexes were less efficiently displaced and displayed a relatively higher stability in the presence of excess histidine, suggesting a trend of coordinating ability of different amino acids towards the 99mTc-MAG3 chelating system. The percentage of 99mTc-peptides remaining intact after incubation with excess histidine ranged from 90 to 97% at 1 h and 88 to 94% at 4 h, versus 70 to 95% at 1 h and 30 to 75% at 4 h with cysteine, clearly demonstrating a high resistance ability of these peptides to histidine displacement and more susceptibility to cysteine transchelation.
Lipophilicity. The lipophilic characteristics of the five compounds investigated are listed in Table II. The lipophilicity was determined by measuring the distribution of 99mTc-peptides between n-octanol and saline solutions. A good correlation was observed between the log P values and the HPLC retention times of the radiopeptides under investigation. The results demonstrated that the 99mTc-MAG3-HER2/neu, which displayed the highest retention time (19.3 min) on HPLC, is the most lipophilic compound (log P=1.31±0.11) evaluated in this study, whereas 99mTc-MAG3-EPPT, with a retention time of 12.0 min, is the most hydrophilic compound (log P=-3.15±0.05). The obtained log P values varied between 1.31±0.11 and -3.15±0.05 (Table II), revealing a variable degree of lipophilicity of the 99mTc-complexes evaluated in this study. The lipophilic characteristic of a radiopeptide is important in predicting its route of excretion since hydrophilic radiopharmaceuticals are generally excreted through the renal-urinary pathway, the preferred route of excretion of an effective tumor-imaging peptide radiopharmaceutical.
In vitro stability in plasma and in T47-D cells. The proteolytic degradation of 99mTc-peptides was determined in vitro in human plasma and also in T47-D cells. Following incubation of the radiolabeled peptides in plasma, the plasma proteins were precipitated with acetonitrile/ethanol, and the samples were analyzed by HPLC (Figure 1). All radiopeptides displayed a significant reformation to free 99mTcO4- (>25%) during incubation in plasma, particularly after 4 h at 37°C. The metabolic stability varied between 40 and 90% at 1 h, and 30 and 87% at 4 h, revealing a fast degradation of radiopeptides by plasma proteases (Table IV). Enzymatic degradation was even quicker with antitumor-antibody-derived peptides, as only 40 and 55% of 99mTc-MAG3-EPPT and 99mTc-MAG3-EPPT-MUC1, respectively, remained intact after 1 h at 37°C. After 4 h these two peptides were almost completely degraded (~70%). Surprisingly, both these metabolically-unstable peptides have shown a high chemical stability against cysteine and histidine challenge for up to 4 h at 37°C (Table III).
99mTc-peptides exhibited a much slower degradation in T47-D cells than in human plasma, as revealed by the HPLC analysis after 1 h incubation with the cells (Table IV). The radiopeptides were found to be relatively stable (≥80%) in the cell culture, an important characteristic, since radioligand stability in the cell culture may be desirable for efficient receptor binding.
In vitro cell binding and internalization. The radiopeptides were evaluated for their tumor targeting properties in vitro using three well-characterized human breast cancer cell lines: estrogen receptor-positive MCF-7, estrogen-independent MDA-MB-231, and T47-D human ductal breast cancer (40). These cell lines presented diverse levels of expression of both HER2/neu and MUC1 (12, 41-45). The binding of peptides to their respective receptor-expressing cell lines was determined by saturation assays (Table V). The data demonstrate that 99mTc-MAG3-HER2/neu showed a similar binding affinity to MCF-7 (7.92±1.09 nM) and T47-D (7.83±1.64 nM) - both these cell lines express moderate levels of HER2/neu (42). Astonishingly, this radiopeptide exhibited a 2-fold higher affinity (3.33±0.92 nM) to MDA-MB-231, despite the fact that this cell line does not have high levels of HER2/neu expression (42, 43). 99mTc-MAG3-MUC1, on the other hand, displayed a high affinity (7.47±1.53 nM) to receptor-positive T47-D (44, 45), but showed a 2-fold lower affinity (14.85±1.83 nM) to the moderate level expressing MCF-7 (44). The radiopeptide displayed a relatively low affinity (38.77±3.13 nM), to a cell line (MDA-MB-231) with low levels of expression for MUC1 (44, 45). The hybrid-tumor-specific peptides were synthesized with the aim of enhancing the binding efficacy by targeting multiple receptor types, but no significant improvement in terms of binding efficiency over their corresponding nonhybrid derivatives was observed. As can be seen from the data (Table V), 99mTc-MAG3-HER2/neu-MUC1 displayed a variable affinity profile as compared to its corresponding nonhybrid derivatives (99mTc-MAG3-HER2/neu and 99mTc-MAG3-MUC1). The antitumor-antibody-derived peptides (99mTc-MAG3-EPPT and 99m Tc-MAG3-EPPT-MUC1) showed similar binding to the MCF-7 cell line, 3.95±0.80 nM vs. 3.36±1.01 nM, respectively, whereas when binding with T47-D and MDA-MB-231, 99mTc-MAG3-EPPT was found to be superior as it displayed a 2- to 5-fold higher affinity when compared with its MUC1-hybrid-derivative. The results obtained from in vitro binding studies confirmed the significant affinity and specificity of the radiopeptides towards their respective receptor-expressing tumor cell lines.
It is important for peptidic radionuclide therapy that the peptides be internalized into tumor cells after binding to cell-surface receptors (34, 36). In vitro internalization studies were performed to determine the extent of internalization of the 99mTc-peptides in human breast cancer cells. In order to differentiate between the tightly cell surface-bound and internalized radioligands, cell-bound radioactivity was treated with acid buffer for 20 min at the end of the binding experiments (46). A rapid and significant internalization was observed for 99mTc-MAG3-HER2/neu and 99m Tc-MAG3-MUC1, with 43.0±2.90% and 30.0±2.50%, respectively, of the cell surface-bound radiopeptides internalized into T47-D cells after incubation in acid buffer for 20 min at 37°C. The percent of specific cell surface-bound 99mTc-peptides internalized into MCF-7 cells varied from 9 to 36% and from 4 to 35% into MDA-MB-231 cells. For T47-D cells, the percent internalized was found to be between 20 and 43% (Table VI). A low internalization (9.0±1.29%) of 99mTc-MAG3-HER2/neu into MCF-7 cells is most likely because of the low levels of HER2/neu expression by MCF-7 cells (12). 99mTc-MAG3-MUC1 showed a fairly low intracellular radioactivity (4.0±0.99%) into MDA-MB-231, which may be associated with a very low MUC1 expression on MDA-MB-231 cells (44, 45). These findings of cell-binding and internalization into tumor cells demonstrate that, despite the modifications in the sequences for labeling and the introduction of a spacer group, these peptides maintained their integrity and held sufficient affinity and specificity towards receptor-positive breast cancer cell lines.
In vivo biodistribution and tumor uptake. Table VI summarizes results of biodistribution studies in normal Balb/c mice at 1 and 4 h p.i. In vivo biodistribution of 99mTc-peptides were performed first in normal mice (without tumors) to determine normal tissue distribution and clearance kinetics. The results demonstrate that all the conjugates revealed a rapid and efficient clearance from the blood at 1 and 4 h p.i. The amount of radioactivity generally retained in the liver ranged from 0.33±0.04% ID/g to 5.83±0.78% ID/g at 1 h, and from 0.29±0.05% ID/g to 4.14±1.71% ID/g at 4 h p.i. The highest uptake was by 99mTc-MAG3-HER2/neu-MUC1 (5.83±0.78% ID/g and 4.14±1.71% ID/g, at 1 and 4 h, respectively) possibly in accordance with its high lipophilicity. Although the radioactivity in the intestines was measured without the contents, still a considerably high uptake was observed (up to 13.33±4.60% ID/g). The uptake in the intestines of different complexes varied between 2.89±1.35% ID/g and 13.33±4.60% ID/g at 1 h p.i. and between 0.94±0.33% ID/g and 6.56±1.38% ID/g at 4 h p.i. The accumulation in the kidneys ranged from 1.02±0.28% ID/g to 3.68±1.27% ID/g at 1 h p.i. and from 0.72±0.15% ID/g to 1.83±0.38% ID/g at 4 h p.i., demonstrating a low uptake/retention of these complexes by the kidneys. The radioactivity accumulation in the stomach for all the peptides (except for 99mTc-MAG3-EPPT) increased over time, with a maximum uptake of 4.0±1.29% ID/g at 4 h p.i., showing in vivo stability of the 99mTc-complexes and insignificant reformation to 99mTcO4-. The radiopeptide 99m Tc-MAG3-EPPT-MUC1, with the highest stomach uptake (4.0±1.29% ID/g, 4 h), also showed a relatively poor stability in human plasma in vitro, signifying a correlation between in vitro and in vivo stability but not with the overall biodistribution and clearance pattern. The uptake in the lungs was also low (values varied from 0.10±0.02% ID/g to 1.30±0.22% ID/g) both at 1 and 4 h p.i. The data suggest that the introduction of a hydrophilic spacer (aspartic acid) only minimally improved renal excretion of the various bioconjugates studied. In general, a rapid and efficient clearance of radioactivity was achieved from all the organs except the intestines, highlighting the favorable biodistribution profiles of these compounds.
It is worth mentioning that precautions were taken to maximize collection of urine, but because of the practical difficulties, complete collection of urine from mice after sacrifice is not always possible, which may result in underestimation of urinary excretion. It can be assumed that the majority of the radioactivity in the intestines arrives via the hepatobiliary system. It was observed that most of the radioactivity was found in the contents of the intestines. The clearance properties of 99mTc-complexes are presented in Figure 2. It is apparent from the clearance patterns in normal mice that the radiopeptides cleared by both the hepatobiliary and renal-urinary pathway, but the degree of clearance from each route varies from peptide to peptide. For instance, 99mTc-MAG3-HER2/neu, the most lipophilic complex studied, with a log P=1.31±0.11, was cleared significantly by both the renal-urinary and the hepatobiliary systems. The clearance pattern of 99mTc-MAG3-HER2/neu-MUC1 was nearly identical to that of the analogous 99mTc-MAG3-HER2/neu. In contrast, 99mTc-MAG3-MUC1 was excreted mainly through the renal-urinary route with some elimination via the hepatobiliary system. The most favorable clearance pattern was observed for 99mTc-MAG3-EPPT-MUC1, with the renal-urinary pathway as the main excretory route (67-68% ID) in accordance with the hydrophilic characteristic (log P=-2.50±0.04) of this radioconjugate. The corresponding 99mTc-MAG3-EPPT also behaved similarly, but the extent of renal-urinary excretion was found to be slightly lower (39-43% ID) than that of 99mTc-MAG3-EPPT-MUC1. The clearance studies underline the importance of the degree of lipophilicity as one of the determining factors in the clearance passage of a radiopharmaceutical.
99mTc-MAG3-HER2/neu, 99mTc-MAG3-MUC1, 99mTc-MAG3-EPPT and 99mTc-MAG3-EPPT-MUC1 were further evaluated in nude mice bearing MDA-MB-231 xenografts to determine their ability to target human breast cancer in vivo (Table VIII). The highest tumor uptake was achieved with the analog 99mTc-MAG3-EPPT-MUC1, 2.20±0.98% ID/g, as early as 1 h p.i., which fell to 0.63±0.21% ID/g at 4 h p.i. Interestingly, the same compound exhibited the most favorable clearance kinetics in normal mice (Table VII). However, an enhanced uptake of this compound in most of the major organs, especially in the kidneys (4.95±1.27% ID/g, 1h), was found in tumor-bearing mice as compared to normal mice. A blocking dose of MAG3-EPPT decreased the uptake of 99m Tc-MAG3-EPPT-MUC1 in the tumor, intestines and kidneys, whereas uptake in the lungs was increased (data not shown). In nude mice bearing breast tumor xenografts, the accumulation in the tumor was moderate for all the four conjugates examined particularly at 1 h p.i. (Table VIII). The tumor uptake values varied from 1.27±0.60% ID/g to 2.20±0.98% ID/g at 1 h and from 0.36±0.11% ID/g to 1.59±0.60% ID/g at 4 h p.i. The radiopeptides showed greater accumulation in the tumors than in the blood and muscle in the MDA-MB-231-bearing nude mice. Though the uptake in tumors was not very high for any of the peptides tested (up to 2.20±0.98% ID/g), slow tumor washout was observed for these peptides and in particular for 99mTc-MAG3-EPPT (1.65±0.64% ID/g at 1 h vs. 1.59±0.60% ID/g at 4 h) and 99mTc-MAG3-HER2/neu (1.60±0.61% ID/g at 1 h vs. 1.03±0.50% ID/g at 4 h). The tumor retention of the two other peptides (99mTc-MAG3-EPPT-MUC1 and 99mTc-MAG3-MUC1) was found to be relatively low (with 71-72% washout).
Radioactivity uptake pattern for most tissues was comparable in nude and normal mice, but in some cases organ uptake was lower in tumor-bearing mice than in normal mice, especially in the intestines. The uptake in the kidneys was low to moderate and ranged between 1.26±0.34% ID/g and 4.95±1.27% ID/g at both time points. Less than 6% ID/g of uptake was obtained for the major organs examined in tumor-bearing mice (Table VIII).
Tumor-to-nontumor ratios were also calculated and are presented in Figure 3. For 99mTc-MAG3-EPPT-MUC1, the tumor-to-blood ratios was approximately 2:1 at 1 h p.i. and 7:1 at 4 h p.i. The relatively low tumor-to-blood ratio at 1 h p.i. is due to the presence of a somewhat high radioactivity in the blood. The tumor-to-muscle ratios of this peptide were approximately 25:1 at 1 h p.i. and 21:1 at 4 h p.i. The other peptides (99mTc-MAG3-HER2/neu, 99m Tc-MAG3-MUC1 and 99mTc-MAG3-EPPT) also displayed fairly good tumor-to-blood ratios varying from 1:1 to 4:1 at 1 h p.i. and from 1:1 to 5:1 at 4 h p.i. The tumor-to-muscle ratios ranged from 17:1 to 21:1 at 1 h p.i. and from 12:1 to 16:1 at 4 h p.i., demonstrating the possible potential of these peptides for targeting breast tumors in vivo. A trend of increased tumor-to-blood ratios and reduction in tumor-to-muscle ratios over time was obtained for these peptides. These results suggest that the peptides evaluated in this study indeed have favorable characteristics which may make them potential candidates for imaging breast cancer in vivo.
The preparation, and in vitro and in vivo evaluation of five new peptides derived from tumor-specific antigens and antitumor-antibody were investigated in this study. The peptides, after radiolabeling with 99mTc via ligand exchange, displayed good chemical and metabolic stability in vitro. The data from in vitro binding studies performed with human breast cancer cell lines presenting diverse levels of receptor expression confirmed the significant affinity, in a low nanomolar range, towards the respective receptor-expressing cell lines. Moreover, the peptides displayed a good degree of internalization into the cells after binding to the cell surface receptors. In vivo biodistribution in Balb/c mice was characterized by an efficient clearance from the blood and a low to moderate uptake/retention of radiopeptides by the major organs (up to 6% ID/g), with the exception of the intestines, which retained up to 13% ID/g. These peptides, with variable degrees of lipophilicity, exhibited different clearance profiles. The hydrophilic peptides, such as 99mTc-MAG3-EPPT-MUC1 and 99mTc-MAG3-MUC1, cleared mainly through the renal-urinary route while the peptide with a higher lipophilicity, 99mTc-MAG3-HER2/neu, was eliminated by both the renal-urinary and hepatobiliary system. In nude mice with breast tumor xenografts, moderate uptake in the tumor was observed with favorable tumor-to-blood and tumor-to-muscle ratios, indicating the tumor targeting potential of these peptides. The data suggest that the peptides under investigation have favorable in vitro and in vivo properties which make them attractive candidates for the detection of breast cancer in vivo. Although more studies are required to determine the full potential of these peptides as breast cancer imaging agents, these findings provide useful information for designing and developing new tumor-specific peptides for targeting tumors that overexpress specific receptors for the peptides.
Acknowledgements
The authors would like to thank Marsood Ahmad for help in animal experiments and Dr. H. Ghebeh for the preparation of cells. This work was supported by a grant from the KACST (Grant No. M-S-12-2 awarded to S.M.O.).
- Received November 3, 2008.
- Revision received January 7, 2009.
- Accepted February 13, 2009.
- Copyright© 2009 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved