Introduction

Cancer is among the most common causes of death throughout the world. In 2008, approximately 13 million new cancer cases and 7.6 million cancer deaths are estimated to have occurred worldwide [1]. Cancer imaging techniques using radiotracers targeted to specific receptors have yielded successful results, demonstrating the utility of such approaches for developing specific radiopharmaceuticals [2–7]. Molecular imaging of tumor metabolism, proliferation, and other-specific targets is a powerful tool in the diagnosis, staging, restaging, response evaluation and guiding surgery, radiotherapy, and systemic treatment [8, 9].

Regulatory peptide receptors are over-expressed in numerous human cancer cells. These receptors have been used as molecular targets for radiolabeled peptides to locate tumors. In recent years, many studies have been performed to identify peptide analogs able to target these tumors, such as gastrin-releasing peptides, somatostatin, neurotensin, and vasoactive intestinal peptides [10–15].

Bombesin (BBN), a tetradecapeptide, was first isolated by Anastasi et al. [16] from the skin of European Bombina bombina frog. The mammalian counterpart is the 27 amino acid gastrin-releasing peptide (GRP). BBN and GRP differ by only 1 of 10 carboxy-terminal residues, which explains the similar biological activity of the two peptides [17–19]. The bombesin receptor family is comprised of four receptor subtypes: (1) neuromedin B receptor (BBN1); (2) gastrin-releasing peptide receptor (BBN2); (3) the orphan receptor subtype (BBN3); and (4) the bombesin receptor subtype (BBN4) [20–22]. A variety of tumors have been found to express receptors for these peptides, such as lung, prostate, breast, pancreas, and colon [23]. Radiolabeled BBN analogs with a high affinity for these receptors might therefore be used for scintigraphic imaging of these tumor types [24–26]. Several of these analogs bind selectively and avidly to GRP receptors on cancer cells when the truncated amino acid sequence (BBN(7–14)NH2) is used. Prior studies have been reported that the C-terminal amino acid sequence is necessary to retain receptor binding affinity. Thus, the N-terminal region of the peptide can be used for radiolabeling [10, 27–31].

Technetium-99m (99mTc) has mostly been used to label radiopharmaceuticals, due to its suitable physical and chemical characteristics and inexpensive isotope cost [32, 33]. 2-Hydrazinonicotinamide (HYNIC) is an attractive bifunctional chelating ligand used to prepare 99mTc-labeled peptides [34], as it shows a high labeling efficiency and its usage with various co-ligands (e.g., ethylendiaminediacetic acid (EDDA), tricine and glucoheptonate) allows for the control of hydrofobicity and pharmacokinetics of the small 99mTc-labeled peptides [35].

Radiolabeling and biodistribution studies of 99mTc-HYNIC-βAla-Bombesina(7–14) can be found in prior studies carried out by the present research group [2], the results of which showed that the complex was able to identify Erhlich tumors, a form of murine breast cancer. Therefore, the purpose of the present study was to demonstrate the ability of 99mTc-HYNIC-βAla-Bombesina(7–14) to identify human breast tumors (MDA-MB-231 cell line) in athymic nu/nu mice. To achieve this, in vitro assays, biodistribution studies were performed, and scintigraphic images were acquired.

Materials and methods

Materials

The peptide HYNIC-βAla-Bombesin(7–14) was purchased from GL Biochem (Shanghai, China). Technetium-99m was obtained from an alumina-based 99Mo/99mTc generator. All solvents (HPLC analytical grade) and other reagents, including tricine, ethylenediamine-N,N′-diacetic acid (EDDA), and SnCl2·2H2O, were purchased from Sigma-Aldrich (São Paulo, Brazil). The subcutaneous tumor model was established in a 6–8 week-old female athymic nu/nu mice purchased from CEMIB (Campinas, Brazil). All animal studies were approved by the local Ethics Committee for Animal Experiments (CETEA).

Radiolabeling

To a sealed vial containing 20 mg tricine and 5 mg of EDDA was added 0.5 ml of 0.9 % NaCl. Next, ten micrograms of HYNIC-βAla-Bombesin(7–14), and 10 μl of 4.5 mM SnCl2·2H2O solution in 0.1 N HCl were added. The pH was adjusted to 7 with 10 μl of NaOH (1 mol/l). Next, an aliquot of 0.5 ml of Na99mTcO4 (37 MBq) was added. The solution was heated for 15 min in a water bath at 100 °C and cooled in water.

Radiochemical purity

Radiochemical purity analyses were performed by instant thin layer chromatography on silica gel (ITLC-SG, Merck) and reverse phase high-performance liquid chromatography (HPLC).

ITLC-SG analysis was accomplished using two different mobile phases: Methylethylketone to determine the amount of free 99mTcO4 − and a solution of acetonitrile:water (1:1) to identify the 99mTcO2. The HPLC analysis was performed using a Waters 717 with a radioactivity detector. HPLC solvents consisted of H2O, containing 0.1 % trifluoroacetic acid (solvent A) and acetonitrile containing 0.1 % trifluoroacetic acid (solvent B). A Symmetry C-18 column (5.0 μm, 4.6 × 150 mm) was used at a rate of 1.0 ml/min. The HPLC gradient system began with a solvent composition of 95 % A and 5 % B, and followed a linear gradient of 30 % A and 70 % B for 10 min and 5 % A: 95 % B from 10 to 15 min. In this system, retention times for free 99mTcO4 − and 99mTc-HYNIC-βAla-Bombesin(7–14) were 2–3 and 5–6 min, respectively.

Partition coefficient

Aliquots of 0.1 ml of the 99mTc-HYNIC-βAla-Bombesin(7–14) (0.37 MBq) were added to tubes containing 2.0 ml of n-octanol/water (1:1). The tubes were vigorously stirred for 3 min. After phase separation, aliquots of 0.5 ml from each phase were collected. Partition coefficient was determined using radioactivity measured in each aliquot by an automatic scintillation apparatus.

Plasma stability

ITLC-SG were used to estimate the plasma stability of 99mTc-HYNIC-βAla-Bombesin(7–14). A volume of 90 μl of labeled peptide solution was incubated, under agitation, at 37 °C with 1.0 ml of fresh mice plasma. Radiochemical stability was determined from samples taken at 1, 2, 4, 6, and 24 h after incubation.

Cysteine challenge

A fresh cysteine solution was prepared and diluted in different concentrations. Next, 0.9 ml of each cysteine solution was mixed with 0.1 ml of the labeled peptide solution. The molar ratios of cysteine to peptide were 1:1, 10:1, and 100:1. Each tube test was incubated at 37 °C, and the radiochemical purity was analyzed by ITLC at 1, 2, 4, 6, and 24 h post-incubation.

Blood clearance

The 99mTc-HYNIC-βAla-Bombesin(7–14) complex was administrated to each mouse (n = 3) through a tail vein, and blood samples (approximately 50 μl each) were collected at 1, 3, 5, 10, 15, 30, 45, 60, 90, and 120 min after administration. A small incision was made in the distal tail to facilitate rapid and reliable blood collection. Each sample was weighed, and the associated radioactivity was determined in an automatic scintillation apparatus. The percentage injected dose per gram (%ID/g) and its mean ± SD in each sample were determined, and the data were plotted as a function of time. Blood clearance analysis was performed using RSTRIP II (Micromath, Salt Lake City, UT, USA).

Cell culture

MDA-MB-231 cells were grown in Dulbecco’s modified Eagle’s medium (D-MEM, Gibco, USA), supplemented by 10 % (v/v) fetal bovine serum, penicillin (100 IU/ml), gentamicin (60 μg/ml), amphotericin B (0.25 μg/l), and streptomycin (100 μg/ml). Cells were kept in humidified air containing 5 % CO2 at 37 °C. The cells were grown to confluence and later harvested by trypsinization. After centrifugation (5 min at 330×g), cells were re-suspended in PBS for inoculation into the athymic nu/nu mice.

Tumor cell inoculation

Aliquots (100 μl) with 1 × 107 MDA-MB-231 cells were subcutaneously injected into the right thigh of female athymic nu/nu mice (17–23 g). Tumor cells were allowed to grow in vivo for 3 to 4 weeks post-inoculation, thus forming tumors with a diameter of no more than 10 mm. Breast tumor-bearing athymic nu/nu mice were used for biodistribution studies and scintigraphic images.

Cell binding, internalization assay, and non-specific binding

MDA-MB-231 cells supplied in D-MEM medium were diluted to 1 × 106 cells/tube and incubated with 99mTc-HYNIC-βAla-Bombesina(7–14) (0.3 nmol total peptide) in triplicate at 37 °C for 1 h and 4 h. The tubes were centrifuged (10 min, 3,000×g) and washed with 0.9 % NaCl. The activity of the cell pellet was determined in an automatic scintillation apparatus. Radioactivity in the cell pellet represents both externalized peptide and internalized peptide. An aliquot with the initial activity was taken as 100 %, and the cell uptake activity was then calculated.

To determine the percentage of internalization, the cell surface-bound radioligand (externalized peptide) was removed using an acid wash buffer (1 ml of 0.2 M acetic acid/0.5 M NaCl; pH 2.8) at room temperature for 5 min. The test tubes were centrifuged, washed with 0.9 % NaCl, and re-centrifuged. Pellet activity represented internalization. Non-specific binding was performed in parallel using the same aforementioned protocol; however, 40 μl (10 μM) of the cold HYNIC-βAla-Bombesin(7–14) were used to block GRP receptors.

Biodistribution studies

Aliquots of 3.7 MBq of the 99mTc-HYNIC-βAla-Bombesin(7–14) were injected intravenously into tumor bearing-athymic nu/nu mice (n = 5). After 1 h and 4 h, mice were anesthetized with a mixture of xylazine (15 mg/kg) and ketamine (80 mg/kg). Whole liver, spleen, kidney, stomach, heart, lungs, blood, muscle, thyroid, intestines, pancreas, and tumor were all removed, washed with distilled water, dried on filter paper, and placed in pre-weighed plastic test tubes. The radioactivity was measured using an automatic scintillation apparatus. A standard dosage containing the same injected amount was counted simultaneously in a separate tube, which was defined as 100 % radioactivity. The results were expressed as the percentage of injected dose/g of tissue (%ID/g). Receptor blocking studies were also carried out by the administration of 40 μg of cold bombesin together with the 99mTc-HYNIC-βAla-Bombesin(7–14).

Scintigraphic images

Aliquots of 18 MBq of the 99mTc-HYNIC-βAla-Bombesin(7–14) were injected intravenously into tumor bearing-athymic nu/nu mice (n = 5). Anesthetized mice were horizontally placed under the collimator of a gamma camera (Mediso, Hungary) employing a low-energy high-resolution collimator. Images were acquired at 1 and 4 h post-injection using a 256 × 256 × 16 matrix size with a 20 % energy window set at 140 keV for a period of 300 s.

Statistical analysis

All data are expressed as mean ± SD. Means between the various groups were compared for differences with analysis of variance. In case of multiple comparisons, a post hoc Bonferroni correction was applied. A P value <0.05 was considered to indicate a statistically significant difference. All data were analyzed by GraphPad PRISM version 5.00 software.

Results and discussion

Radiochemical purity and partition coefficient

The ITLC and HPLC were used for radiochemical analyses to predict the radiochemical purity of the 99mTc-HYNIC-βAla-Bombesina(7–14) complex, as previously described [2]. The results obtained by ITLC analyses showed a mean radiochemical purity for 99mTc-HYNIC-βAla-Bombesina(7–14) of 97.8 ± 0.9 % (n > 15), which remains stable after 24 h without post-labeling purification. HPLC analysis showed similar results, the 99mTc-HYNIC-βAla-Bombesina(7–14) complex presented a retention time of 5.26 min, and the lower peak, in a retention time of 2.41 min, was considered to be pertechnetate (Fig. 1, black line). The area for both peaks were calculated, and the 99mTc-HYNIC-βAla-Bombesina(7–14) complex showed a radiochemical purity equal to 96.2 %. In addition, HPLC analysis of the 99mTc-HYNIC-βAla-Bombesina(7–14) complex revealed high stability, since the complex showed the same retention time (Fig. 1, gray line) after 24 h post-labeling. The difference of intensity between the post-labeling line (black line) and the 24 h post-labeling line was attributed to the decay of technetium-99m. The presence of radiochemical impurities proved to be a drawback in nuclear medicine, yielding images with poor quality. Thus, it can be concluded that radiopharmaceuticals should contain high radiochemical purity (above 90 %). Therefore, the 99mTc-HYNIC-βAla-Bombesina(7–14) complex presented good chemical characteristics, since it presented a high radiochemical purity (>95 %).

Fig. 1
figure 1

Reverse phase HPLC radiochromatograms of 99mTc-HYNIC-βAla-Bombesina(7–14) immediately after labeling and 24 h after preparation (kept at room temperature)

Full size image

Partition coefficient of the radiolabeled peptide was determined by the ratio between n-octanol and water. log P of 99mTc-HYNIC-βAla-Bombesina(7–14) could be observed within the hydrophilic range (log P = −1.78).

Plasma stability and cysteine challenge

Radiochemical stability was evaluated for the 99mTc-HYNIC-βAla-Bombesina(7–14) radiopharmaceutical at 1, 2, 4, 6, and 24 h in mice plasma. An excellent stability, even over long periods of time (>90 %), could be observed. The radiopharmaceutical was also evaluated by transchelation toward cysteine. After incubation with 1:1, 10:1 and 100:1 molar ratios of cysteine to peptide, ITLC revealed that the radioactivity dissociated from 99mTc-HYNIC-βAla-Bombesina(7–14) was less than 10 % (Table 1). These data indicated a high stability in all assays.

Table 1 Cysteine tranchelation (% radiochemical purity)
Full size table

Blood clearance

Blood clearance for 99mTc-HYNIC-βAla-Bombesina(7–14) was rapid (Fig. 2), showing a biphasic profile with α half-life of 12.9 min and β half-life of 3.2 min. This data supported early imaging, since the background radiation will not contribute to a decrease in imaging quality [36].

Fig. 2
figure 2

Blood clearance of 99mTc-HYNIC-βAla-Bombesin(7–14) in healthy mice (n = 3)

Full size image

Cell binding and internalization assay

The binding affinity of 99mTc-HYNIC-βAla-Bombesina(7–14) for GRP receptors was evaluated for MDA-MB-231 cells. The in vitro results showed an important uptake, which was significantly inhibited by the co-incubation of cold HYNIC-βAla-Bombesina(7–14) during all assayed times (Fig. 3). This data confirmed the in vitro specificity of 99mTc-HYNIC-βAla-Bombesina(7–14) for GRP receptors presented on cell membranes of the breast tumor.

Fig. 3
figure 3

Uptake of 99mTc-HYNIC-βAla-Bombesina(7–14) by MDA-MB-231 cells. Results are expressed as mean ± standard error

Full size image

Internalization analysis showed moderate values for 99mTc-HYNIC-βAla-Bombesina(7–14), presenting 16.8 and 20.0 % internalized activity after 1 and 4 h post-injection, respectively. The values increased proportionally with the time of incubation, illustrating that internalization was time-dependent. These results were higher than values reported for other BBN analogs [13], which confirmed that this analog presents a sufficient affinity for GRP receptor-expressing tumors.

Biodistribution studies and scintigraphic images

Results obtained from in vitro studies suggested that the BBN derivative can be used as a radiotracer. However, these tests are not a dependable index of clinical usefulness. Therefore, it is only possible to predict the real feasibility of 99mTc-HYNIC-βAla-Bombesina(7–14) after having performed biodistribution studies [37].

99mTc-HYNIC-βAla-Bombesina(7–14) biodistribution is shown in Table 2. Maximum uptake could be observed in the kidneys, indicating a main renal excretion; however, hepatobiliary clearance is also present. These data corroborate with partition coefficient results, since hydrophilic molecules are preferably eliminated through renal excretion [38]. The radiotracer revealed rapid blood clearance, with only 0.45 %ID/g at 1 h, followed by a further decrease at 4 h. This result coincides with data shown in the blood clearance assay, which is important to allow for early imaging [36].

Table 2 Biodistribution (blocked and unblocked) of 99mTc-HYNIC-βAla-Bombesina(7–14) in breast tumor-bearing athymic nu/nu mice
Full size table

The pancreas is typically used as an indicator of GRP receptor specificity for BBN derivatives [37, 39]. In the present study, the pancreas presented a higher uptake, when compared with the non-excretory organs, such as the spleen, the liver, and muscles due to its GRPR expression, indicating that BBN acts as a targeting vector. Moreover, in blocked studies, pancreas uptake at 1 and 4 h post-injection was reduced, from unblocked biodistribution, by 56 and 87 %, respectively (Table 2).

This finding suggests specificity for GRP receptors.

Tumors showed a moderate uptake, the highest of which was recorded at 4 h, indicating the trapping of 99mTc-HYNIC-βAla-Bombesin(7–14) in the tumor site, due to the fact that GRP receptors are present on the surface of MDA-MB-231 cell line [10, 12, 40]. Higher tumor uptakes have been reported in other studies [41–43]; however, this can be explained due to the fact that different cell lines were evaluated (e.g., PC-3), which express higher GRP receptor densities when compared to the cell lines analyzed in this work (MDA-MB-231) [13, 44]. Nevertheless, tumor-to-muscle and tumor-to-blood ratios for 99mTc-HYNIC-βAla-Bombesin(7–14) presented high values at 4 h (6.00 and 4.69, respectively). It has been considered in the literature [45] that radiotracers showing target/non-target ratios of greater than 1.5 (50 % higher uptake in the target tissue) may be considered potential diagnostic agents. Furthermore, when cold HYNIC-βAla-Bombesin(7–14) was co-administrated, the results showed a significant reduction in tumor-to-muscle ratios of 50.0 % at 1 h and 65.2 % at 4 h post-injection, as observed in Fig. 4. Therefore, there is strong evidence pointing to the 99mTc-HYNIC-βAla-Bombesin(7–14) specificity for GRP receptors.

Fig. 4
figure 4

Tumor-to-muscle ratios after the injection of 99mTc-HYNIC-βAla-Bombesin(7–14) in breast tumor-bearing athymic nu/nu mice. (n = 5). Results are expressed as mean ± standard error

Full size image

Scintigraphic studies performed in breast tumor-bearing athymic nu/nu mice revealed similar excretion profiles. Although tumor uptake was lower than those observed in other studies [41–43], strong signals in the tumor area could be observed in the scintigraphic images, most likely due to the rapid clearance presented for 99mTc-HYNIC-βAla-Bombesin(7–14) (Fig. 5). The quantitative analyses of scintigraphic images showed tumor-to-muscle ratios for 99mTc-HYNIC-βAla-Bombesina(7–14) of 3.56 at 1 h and 5.15 at 4 h, demonstrating no statistical difference between biodistribution and scintigraphic studies. These results showed a tropism of the 99mTc-HYNIC-βAla-Bombesin(7–14) to the tumor during the entire experiment.

Fig. 5
figure 5

Scintigraphic images of breast tumor-bearing athymic nu/nu mice at 1 h (a) and 4 h (b) after radiopharmaceutical administration. While under ketamine/xylazine anesthesia, 18 MBq of 99mTc-HYNIC-βAla-Bombesina(7–14) were injected into the tail vein. The length of scan was 300 s. The arrows show tumor area

Full size image

Conclusions

The peptide HYNIC-βAla-Bombesin(7–14) was successfully labeled with technetium and demonstrated a high level of stability. In vitro assays confirmed the affinity of 99mTc- HYNIC-βAla-Bombesin(7–14) to MDA-MB-231 cells. Biodistribution and scintigraphic studies showed adequate tumor-to-muscle ratios in breast tumor-bearing athymic nu/nu mice. In summary, these results showed the feasibility of 99mTc- HYNIC-βAla-Bombesin(7–14) as a functional agent in tumor diagnoses.