Tc-Based Amyloid Probes

Bioconjugate Chem. 2008, 19, 1087–1094

1087

Synthesis and Screening of a Library of Re/Tc-Based Amyloid Probes Derived from β-Breaker Peptides Karin A. Stephenson,† Leslie C. Reid,† Jon Zubieta,‡ John W. Babich,§ Mei-Ping Kung,⊥ Hank F. Kung,⊥ and John F. Valliant*,† Departments of Chemistry and Medical Physics & Applied Radiation Sciences, McMaster University, Hamilton, ON, Canada, L8S 4M1, Department of Chemistry, Syracuse University, Syracuse, New York 13244-4100, Molecular Insight Pharmaceuticals Inc., Cambridge, Massachusetts 02142, and Department of Radiology, University of Pennsylvania, Philadelphia, Pennsylvania 19104. Received January 2, 2008; Revised Manuscript Received February 13, 2008

Through the development and application of a unique approach for producing Re-metallopeptides, a new class of peptide-derived probes that are designed to target β-amyloid plaques was developed. Derivatives of a class of β-breaker peptides having the core sequence lvffa or affvl (lower case letters represent D-amino acids) and the single amino acid chelate quinoline (SAACQ) ligand which can bind Re and 99mTc were prepared on an automated peptide synthesizer. Both monomeric and dimeric peptides were synthesized in modest to good yields where in select examples a biotin-containing amino acid derivative was included to act as a linker point for further conjugation to carrier proteins. The Re complexes for all reported peptides were prepared similarly and screened for their ability to inhibit fibrillogenesis. Two of the reported compounds showed excellent inhibitory properties (8a: 40 ( 5% amyloid formation versus control; 16: 40 ( 4%) and warrant further investigation. For one of these leads, the 99mTc analogue was synthesized and the product showed high stability toward histidine and cysteine challenges, making it a viable candidate for in ViVo biodistribution studies.

INTRODUCTION Alzheimer’s disease (AD) is a progressive neurodegenerative affliction that involves the formation of plaques and neurofibrillary tangles, which ultimately lead to neuronal death in both the hippocampus and frontal cortex (1, 2). Senile plaques can occupy up to 15% of the brain volume in some advanced AD patients and are composed primarily of β-amyloid peptides (Aβ1-40,42,43) (1, 3). Amyloid peptides are naturally occurring entities, but in AD patients, the peptides form plaques in the brain: a process that stems from a change in the mechanism of cleavage of the amyloid precursor protein (1, 4, 5). The formation of senile plaques has been correlated with the progression of the disease and has therefore emerged as a promising target for the treatment of AD (1, 6, 7). There has been a major effort put into developing radioimaging agents that allow for serial monitoring of the progression of AD and for assessing the impact of therapy. 18F-deoxyglucose (FDG) is used to diagnose the early onset of disease by visualizing reduced metabolism of glucose associated with cell death in the brains of AD patients (2). Unfortunately, FDG is not able to specifically identify β-amyloid plaques; therefore, to better characterize changes in plaque load, work in the field has been geared toward developing plaque-targeting PET and SPECT agents (8-12). Maggio and co-workers reported that Aβ1-40 labeled with 125 I rapidly deposits on β-amyloid plaques in isolated tissue. Not surprisingly, the labeled peptide showed particularly poor brain uptake in ViVo (13). However, conjugation of an Aβ1-40biotin derivative to a streptavidin-tagged monoclonal antibody * To whom correspondence should be addressed. Tel: 905-525-9140, ext. 22840. FAX: 905-522-2509. E-mail: [email protected]. † McMaster University. ‡ Syracuse University. § Molecular Insight Pharmaceuticals Inc. ⊥ University of Pennsylvania.

(mAb) for the human insulin receptor markedly improved transport across the BBB (14, 15). Uptake of the mAb-125Ibiotin-Aβ1-40 conjugate in rhesus monkey brains was observed using autoradiography. Structure-activity relationships (SARs) have shown that fragments of the full Aβ1-40 peptide can exhibit binding to plaques and inhibition of β-amyloid fibrillogenesis (4, 16-21). β-sheet breaker peptides having as few as four amino acids (LVFF) have been able to significantly inhibit fibril formation (22). The key hydrophobic motif, LVFF, which is common to most AD β-sheet breaker peptides, binds amyloid plaques and can prevent soluble Aβ peptides from forming structured aggregates (4, 16-19). Given that strategies for enabling peptides to cross the BBB exist, radiotracers based on β-breaker peptides would offer a new means for imaging β-amyloid plaques and for assessing therapies aimed at solubilizing plaques. The traditional approach for developing 99mTc-labeled probes of this type would involve appending a technetium chelator to the N-terminus of a core β-breaker peptide sequence. After isolation, attempts would be made to prepare the rhenium complex of the product, which would serve as a reference standard for the 99mTc analogue and as a compound to test the ability of the metallopeptide derivative to bind plaques and/or inhibit amyloid fibrillogenesis. This linear discovery approach is time-consuming with respect to isolating and fully characterizing the Re complexes and identifying the optimal sequence for the vector and site of derivatization on the peptide backbone. We report here a new approach in which an organometallic complex of an amino acid analogue is used to insert the metal into various positions in lvffa and affvl sequences. A synthetic methodology was developed such that multivalent (i.e., dimeric) peptides could be prepared and biotin incorporated into the backbone of the peptide as a linker site for conjugation to carrier proteins. The ability of the Re complexes to prevent fibril

10.1021/bc800001g CCC: $40.75  2008 American Chemical Society Published on Web 04/12/2008

1088 Bioconjugate Chem., Vol. 19, No. 5, 2008

formation was subsequently determined in Vitro in order to identify lead compounds for further testing.

EXPERIMENTAL PROCEDURES Instrumentation. All peptides were prepared on an Advanced ChemTech 348Ω peptide synthesizer using a 40-well reaction vessel. Peptides were analyzed by electrospray mass spectrometry using a Micromass Quattro Ultima instrument in positive ion mode. Samples were dissolved in 50% CH3CN /H2O prior to analysis. FTIR spectra were acquired on a Bio-Rad FTS-40 FTIR spectrometer. Analytical HPLC was performed using a Varian Pro Star model 330 PDA detector, model 230 solvent delivery system, and a C-18 Microsorb column (4.6 × 250 mm, 300 Å-5 µm). Semipreparative HPLC was carried out using a Varian Pro Star HPLC fitted with a model 320 UV detector, a model 215 solvent delivery system, and a C-18 Microsorb column (10 × 250 mm, 300 Å-5 µm). Peptides (1-16) were analyzed using a gradient of 5% to 100% CH3CN containing 0.1% TFA over 40 min at flow rate of 1.0 mL min-1. All runs were monitored at λ ) 254 and 214 nm. For the reactions involving 99mTc, HPLC experiments were performed on a Varian Prostar model 230 HPLC instrument coupled to a Bio-Rad IN/US γ-detector using a Varian Nucleosil (L × ID ) 250 × 4.6 mm) analytical column (300 Å - 5 µm, RP-C18). The mobile phase for the analysis of [Tc(CO)3(H2O)3]+ consisted of solvent A, triethylammonium phosphate buffer (pH ) 2-2.5); solvent B, CH3OH. Gradient: 0-3 min, 100% A; 3-6 min, 100% A to 75% A; 6-9 min, 75% A to 67% A; 9-20 min, 67% A to 0% A; 20-22 min, 0% A; 22-25 min, 0% A to 100% A; 25-30 min, 100% A. For the labeled peptide, the elution conditions were 5-100% CH3CN containing 0.1% TFA over 30 min at flow rate of 1.0 mL min-1. Materials and Methods. Unless otherwise stated, all reagents and solvents were ACS grade or higher and used without further purification from commercial suppliers. Polystyrene-based N-R9-fluorenylmethoxycarbonyl (Fmoc)-glycine loaded Wang resin (0.82 mmol g-1, 1% divinylbenzene, 200-400 mesh) was obtained from NovaBiochem Inc. Fmoc-protected amino acids were purchased from NovaBiochem Inc., Bachem Inc., and Advanced ChemTech Inc. Peptides were prepared on an Advanced Chemtech 348 synthesizer. The Aβ1-40 was purchased from rPeptide (cat# A-1001-1, lot# 3120440A). Caution: 99mTc is radioactive and should only be handled in an appropriately equipped and licensed facility. Solid-Phase Peptide Synthesis. Fmoc-glycine loaded Wang resin (0.82 mmol/g; 100 mg/well) was added to 16 wells in the reaction block, washed with DMF (2 mL/well), and shaken at 600 rpm for 1 min. The wells were then filtered and the resin washed with THF (2 mL/well) three times, shaking at 600 rpm for 1 min and draining for 90 s per wash. The same washing was repeated two additional times with DMF to complete the general wash cycle. This general wash cycle was used between every deprotection and coupling step. Fmoc deprotection was accomplished by adding a 20% v/v piperidine-DMF solution (2 mL/well) and shaking for 5 min at 600 rpm. After draining the reaction wells, the deprotection procedure was repeated this time with shaking for 10 min. The deprotected resin-bound amino acid was then washed with the general wash cycle. For amino acid coupling reactions, standard methods involving HBTU were employed. The linear peptides were prepared by adding DMF (200 µL) to the active vessels followed by a 4-fold excess of the appropriate protected amino acid as a 0.5 M solution in DMF. Four equivalents of HBTU as a 0.5 M solution in DMF was then added, followed by an 8-fold excess of DIPEA as 2.0 M solution in DMF. The reaction vessel was subsequently shaken for 80 min at 600 rpm.

Stephenson et al.

Following filtration, the resin was washed as described above and the cycles repeated to prepare peptides 1-8a. The branched peptides were prepared using the same HBTU coupling strategy described above. At the point of branching, Fmoc-D-Lys(Mtt)-OH was introduced into the peptide where, prior to Mtt deprotection, THF (2 mL) was added to the appropriate reaction wells and the suspension shaken at 600 rpm for 4 min. After filtration, the THF wash was repeated twice to ensure that no residual DMF was present. The Mtt group (Mtt ) 4-methyltrityl) was selectively removed by the manual addition of 2 mL of a freshly prepared cocktail of 1% TFA and 3% TIS in DCM (v/v/v), followed by shaking for 15 min at 600 rpm. After filtration, the procedure was repeated, this time shaking for 30 min. The resin was then tested for complete deprotection by adding a small amount of resin to a 1:1 (v/v) solution of TFA and DCM. When the resin did not turn yellow, Mtt deprotection was assumed to be complete, and the next amino acid was added in 8-fold excess. Eight equivalents of HBTU as a 0.5 M solution in DMF and a 16-fold excess of DIPEA as a 1.0 M solution in DMF were added and the mixture shaken at 600 rpm for 80 min, then filtered. The same conditions were used to add each subsequent amino acid in compounds 9-16. At the end of the synthesis, resin-bound peptides were removed from the reaction block and transferred to 20 mL glass vials. EDT (2%), water (2%), and TIS (2%) in TFA (v/v/v/v) cooled to 0 °C were added to the resin samples and the vials shaken at 400 rpm for 90 min at room temperature. The reaction vessels were subsequently filtered into cold diethyl ether and the heterogeneous mixtures centrifuged at 3000 rpm and -5 °C for 10 min. The resultant pellets were washed with cold diethyl ether (3 × 25 mL), dissolved in distilled-deionized water, and lyophilized to yield white solids for the peptides containing the free ligand and pink solids if rhenium was present (Table 1). Preparation of [99mTc(CO)3(OH2)3]+. A 10 mL penicillin vial fitted with a rubber septum and containing K2[BH3 · CO2], (8.5 mg, 63 µmol), Na2B4O7 · 10H2O (2.9 mg, 8.0 µmol), Na/Ktartrate (15.0 mg, 53 µmol), and Na2CO3 (4.0 mg, 38 µmol) was flushed with N2(g) for 15 min. 99mTcO4- (370 MBq) in 500 µL of saline was added by syringe, and the solution heated at 95 °C for 30 min. After cooling in an ice bath, the solution was neutralized by the addition of 12 M HCl. Gradient HPLC was performed to assess purity and yield. Radiolabeling of Compound 7. Compound 7 (1.0 mg, 0.71 µmol) in 200 µL of distilled-deionized water and 300 µL of CH3CN was added via a syringe to the solution containing [99mTc(CO)3(OH2)3]+ (346 MBq) and the mixture heated to 55 °C for 3 h. The reaction mixture was subsequently loaded onto a solid-phase extraction cartridge (Waters, C18), which had been conditioned with absolute ethanol (10 mL), CH3CN (10 mL), 1:1 CH3CN /10 mM HCl (10 mL), and 10 mM HCl (10 mL) prior to use. Small (1 mL) fractions were collected from the following gradient elution protocol: 10 mM HCl (7 × 1 mL), 1:4 CH3CN/10 mM HCl (2 × 1 mL), 1:1 CH3CN/10 mM HCl (2 × 1 mL), 4:1 CH3CN/10 mM HCl (2 × 1 mL), and finally CH3CN (5 × 1 mL). Compound 8b eluted in fractions 11 and 12, which were concentrated to 200 µL, and the product was isolated by HPLC. The radiochemical yield of 8b was 25%, and the radiochemical purity was >98%. Ligand Challenge Experiments. Two vials each containing 3.7-7.4 MBq of purified 8b in distilled-deionized water (100 µL) were incubated separately with 500 µL of solutions containing cysteine (0.010 M in 100 mM sodium phosphate buffer, pH ) 7.2) and histidine (0.010 M in 100 mM sodium

Re/Tc Amyloid Probes from β-Breaker Peptides

Bioconjugate Chem., Vol. 19, No. 5, 2008 1089

Table 1. Numbering Scheme, Isolated Yield, and Characterization Data for Compounds 1-16a,b compound no.

peptide sequence

yield (%)

HPLC tR (min.)

MS (m/z)

IR υ [M(CO)] (cm-1)

1 2 3 4 5 6 7 8a 9 10 11 12 13 14 15 16

SAACQ-lvffag ReSAACQ-lvffag k(N
Biotin)-SAACQ-lvffag k(N
Biotin)-ReSAACQ-lvffag SAACQ-affvlg ReSAACQ-affvlg k(N
Biotin)-SAACQ-affvlg k(N
Biotin)-ReSAACQ-affvlg lvffa-SAACQ-k(N
affvl)g lvffa-ReSAACQ-k(N
affvl)g lvffa-SAACQ-k(N
affvl)k(N
Biotin)g lvffa-ReSAACQ-k(N
affvl)k(N
Biotin)g affvl-SAACQ-k(N
lvffa)g affvl-ReSAACQ-k(N
lvffa)g affvl-SAACQ-k(N
lvffa)k(N
Biotin)g affvl-ReSAACQ-k(N
lvffa)k(N
Biotin)g

36 63 73 48 66 26 60 72 62 20 31 37 38 37 57 16

20.3 22.3 20.3 22.1 21.0 22.9 20.6 22.7 21.7 23.8 21.0 22.7 22.0 24.0 21.6 23.2

1063 1335 1419 1688 1064 1333 1418 1688 1770 2041 2126b 2394b 1770 2041b 2124b 2395b

– 2031, 1915 – 2031, 1918 – 2031, 1915 – 2032, 1929 – 2032, 1927 – 2032, 1925 – 2031, 1918 – 2031, 1923

a Peptides containing a single targeting sequence are written with the N-terminal group on the left. For the dimeric peptides, the first sequence is written with the N-terminus on the left, while the second group is written beginning with the amino acid attached to lysine at the branching point. b Indicates peak in the MS appears as m/2 and m/3 in positive ion mode.

Table 2. Percentage of Amyloid Fibrils Formed after Incubation of Aβ1-40 with the Re-Containing Peptides at 37 °C for 5 Days compound no.

sequence

% amyloid

2 4 6 8a 10 12 14 16

DMSO control ReSAACQ-lvffag k(N
Biotin)-ReSAACQ-lvffag ReSAACQ-affvlg k(N
Biotin)-ReSAACQ-affvlg lvffa-ReSAACQ-k(N
affvl)g lvffa-ReSAACQ-k(N
affvl)k(N
Biotin)g affvl-ReSAACQ-k(N
lvffa)g affvl-ReSAACQ-k(N
lvffa)k(N
Biotin)g

100 48 ( 4 43 ( 2 47 ( 7 40 ( 5 69 ( 4 55 ( 5 58 ( 4 40 ( 4

phosphate buffer, pH ) 7.2). The reaction mixtures were heated to 37 °C for 24 h after which aliquots were taken and analyzed by HPLC. In Vitro Screening. Aβ1-40 (0.5 mg, rPeptide) was initially dissolved in 100 µL of 0.1% TFA and then 250 µL of 10 mM EDTA added (solution A). The presence of premature fibril formation was checked routinely by taking an aliquot of the solution (10 µL) and adding it to 500 µL of 2 µM Thioflavin T (ThT). The fluorescence was then measured at excitation and emission wavelengths of 440 and 485 nm, respectively, using 10 nm band-pass filters. After ensuring that no fibrils were present, an additional 250 µL of 20 mM sodium phosphate buffer (pH 7.4)/10 mM EDTA was added to solution A to form the Aβ1-40 stock solution (solution B). Rhenium-labeled peptides (2, 4, 6, 8a, 10, 12, 14, and 16) were then dissolved in DMSO and diluted with 20 mM sodium phosphate buffer (pH 7.4) so that the final DMSO concentration was 25%. The required volume of rhenium peptide solution was then added to 40 µL of the Aβ1-40 stock solution (solution B) so that there was a 1:1 molar ratio of Aβ1-40 peptide and the desired rhenium peptide. The resulting solutions were then incubated for 5 days at 37 °C in duplicate. Following incubation, 40 µL of each solution was added to 500 µL of 2 µM ThT and fluorescence measured using a PerkinElmer LS55 luminescence spectrometer (10 nm slit widths). The percentage of amyloid (Table 2) was then calculated by dividing the fluorescence measurements by those found for the control consisting of 25% DMSO in 20 mM sodium phosphate buffer (pH 7.4)/10 mM EDTA and 40 µL of the Aβ1-40 stock solution (solution B).

RESULTS The target peptides were prepared containing the single amino acid chelate-quinoline (SAACQ) ligand, which is an amino acid analogue that is capable of binding Tc and Re. The reported

compounds were derived from the targeting sequence lvffa and the reverse sequence affvl, which are both reported to bind to β-amyloid plaques. The library consisted of 16 unique compounds, half of which possess a single targeting vector (Figure 1), while the remainder possess two targeting sequences (Figure 2). Biotin was introduced into select peptides to provide a convenient means to conjugate the peptides to vectors that facilitate crossing the BBB. For compounds 1-8a, biotin and the SAACQ ligand were incorporated adjacent to the N-terminus of each peptide, as this site is known to tolerate introduction of large groups without affecting binding (4, 16-19), while for compounds 9-16, they were incorporated adjacent to the terminus that is distant to the two targeting sequences. Peptides 1-16 (Table 1) were prepared in parallel on an automated 348 ACT synthesizer using a 40-well reaction block using all D-amino acids. Starting with Fmoc-glycine loaded Wang resin, peptides 1-8a, which contain a single targeting sequence, were synthesized using standard Fmoc/HBTU peptide coupling and deprotection strategies (Scheme 1). In cases where biotin was introduced into the peptides, Fmoc-D-Lys(N
-Biotin)OH was added to the N-terminal amino acid again using standard coupling procedures. Preparation of peptides containing two targeting sequences was accomplished using Fmoc-D-Lys(N
-Mtt)-OH (Mtt ) 4-methyltrityl) as a bridging site (Scheme 2). Mtt was selectively removed during peptide synthesis using a solution of 1% TFA and 3% TIS in DCM (v/v/v). The time required for complete deprotection was monitored by adding a small amount of resin to a 1:1 (v/v) solution of TFA and DCM. In the case where a yellow color appeared, which suggested the presence of Mtt, the reaction was allowed to continue (23, 24). Following complete Mtt deprotection, the Fmoc group was then removed using 20% piperidine/DMF and the two targeting sequences built up simultaneously using an 8-fold excess of the appropriate Fmoc-protected amino acid derivatives per coupling step. Conjugates 1-16 were cleaved from the solid support using a standard cleavage cocktail, precipitated from cold diethyl ether and collected by centrifugation (-5 °C, 10 min, 3000 rpm). Peptides were washed three times with cold diethyl ether, dissolved in water, and lyophilized to yield white solids. HPLC and mass spectrometry analysis confirmed the purity and molecular weight of the peptides (Table 1). In Vitro ScreeningsThioflavin T Inhibition Assay. The Recontaining peptides 2, 4, 6, 8a, 10, 12, 14, and 16 were incubated with the native peptide Aβ1-40 in a 1:1 ratio (10 µM) at 37 °C for 5 days in parallel with controls containing no competitor and 25% DMSO. It was found that 25% DMSO was required

1090 Bioconjugate Chem., Vol. 19, No. 5, 2008

Stephenson et al.

Figure 1. General structure for agents containing a single targeting sequence and biotin (k(N
Biotin)-ReSAACQ-lvffag).

Figure 2. General structure for agents containing two identical targeting sequences and biotin (lvffa-ReSAACQ-k(N
affvl)k(N
Biotin)g).

in order to solubilize the peptide conjugates at the required concentration. Consequently, this solvent composition was used as the reference solution to calculate the percentage of amyloid. CH3CN was initially considered, but test experiments demonstrated that the presence of this solvent increased the formation of fibrils when compared to controls. Following incubation with the peptides, 40 µL aliquots of each mixture were added to 500 µL of 2 µM ThT (giving a final DMSO content that was less than 2%) and the fluorescence intensity measured at excitation and emission wavelengths of 440 and 485 nm, respectively. The percentages of amyloid present (Table 2) were found by dividing the fluorescence intensities observed in the presence and absence of the competitor. Therefore, 100% amyloid would indicate that the peptide was unable to inhibit the formation of fibrils and would correspond to the largest fluorescence intensity. Technetium Labeling. To demonstrate that sufficient quantities of the radiolabeled peptides could be produced for in ViVo imaging studies, compound 7 in a mixture of water and CH3CN was added to a solution containing [99mTc(CO)3(OH2)3]+ and the mixture heated at 55 °C for 3 h (Scheme 3). The labeled product 8b was isolated by solid-phase extraction followed by semipreparative HPLC in 25% radiochemical yield in greater than 98% radiochemical purity. The retention time of 8b in the γ-HPLC matched that for the Re reference standard 8a, which was detected by UV (Figure 3). Cysteine and histidine challenge studies were performed to test the robustness of the metal complex. A large excess of each amino acid was added to solutions of 8b and the mixtures heated

at 37 °C for 24 h. The 99mTc complex showed only very minor signs of degradation (less than 2%) in either test (Figure 4).

DISCUSSION β-breaker peptides are under investigation as an experimental therapeutic strategy for reducing amyloid plaque loads. Radioactive analogues labeled with 99mTc would afford the opportunity to study the distribution of variants of this class of compounds in ViVo, and if they are capable of targeting amyloid plaques, the tracers could also be used to assess the impact of other emerging plaque-reducing treatment strategies. The advantages to using 99mTc over other isotopes are that it is widely available at a low cost and the vast majority of hospitals have the capacity to image Tc labeled compounds. Our group recently described a new and robust method for derivatizing peptides with a chelating system (referred to as the single amino acid chelate quinoline or SAACQ) that is capable of binding 99mTc. The approach is flexible in that the chelate, which is derived from lysine, can be introduced at any position within the backbone of the peptide. A further advantage is that the chelate forms a robust metal complex with rhenium, which can also be incorporated into peptides. As a result, it is conceivable that libraries of all manners of chelate-peptide conjugates and the nonradioactive Re analogues of the target 99m Tc complexes can be prepared in parallel. This would make it possible to screen an exact and nonradioactive replica of any peptide-targeted agent in Vitro prior to undertaking an evaluation of the 99mTc analogue in ViVo.

Re/Tc Amyloid Probes from β-Breaker Peptides

Bioconjugate Chem., Vol. 19, No. 5, 2008 1091

Scheme 1. Synthesis of Single Vector Peptides (3, 4) Containing Biotin

The peptide conjugates 1-16 were successfully prepared in parallel using standard peptide coupling strategies on an automated synthesizer. The peptides were constructed from an Fmoc-glycine loaded Wang resin using D-amino acids, which were employed to avoid future issues of premature degradation of the targeting agent in ViVo. Peptides 1-8a were prepared with a single copy of the lvffa or affvl sequences, while compounds 9-16 contained two copies. The introduction of the two vectors was made possible through the incorporation of Fmoc-D-Lys(Mtt)-OH, which provided a branching point. The preparation of multivalent peptides containing a single chelate group in the backbone of the probe as opposed to conjugated to the N-terminus or a pendent lysine is not commonly observed in radiopharmaceutical chemistry because the appropriately functionalized chelates have not been available until recently. For the bivalent peptides presented here, two identical targeting sequences were used. However, it is easy to see how two unique vectors could be introduced by removing the Mtt group after the first peptide has been synthesized. This type of approach could be used to incorporate a second sequence to improve affinity for the target or to alter the pharmacokinetics. All peptide conjugates (1-16) were cleaved using a standard cleavage cocktail, and the conjugates were isolated in modest to good yields (Table 2) as white solids if they contained SAACQ and as pink solids if they contained Re. HPLC analysis indicated that both monomers and dimers containing SAACQ were prepared such that they did not require further purification. Analogues prepared with ReSAACQ were isolated in greater than 90% purity whereupon HPLC was used to increase this value to 99%. Despite the structural complexity of peptides,

particularly compounds 9-16, it was possible to prepare multimilligram quantities (greater than 50 mg) of each of the products. The ability to scale up reactions, which can be readily accomplished using the reported approach, is an important consideration for compounds that are going to move on to preclinical and clinical evaluation. Peptide conjugates 2, 4, 6, 8a, 10, 12, 14, and 16 were tested for their ability to inhibit fibrillogenesis using a standard assay for determining the ability of compounds, particularly peptides, to bind to β-amyloid plaques (4, 17, 20, 25). Thioflavin T (ThT) is a dye used to stain amyloid fibrils found in the brains of AD patients postmortem (26). When ThT binds to plaques, there is a unique red shift in its absorption spectrum, which is distinct from the unbound dye. The unbound dye has excitation and emission maxima of 385 and 445 nm, which shift to 450 and 485 nm, respectively, upon binding (17, 27). As a result, the formation of ordered Aβ1-40 fibrils can be quantitated in Vitro by measuring the increase in fluorescence intensity at the different wavelengths (20, 28, 29). β-breaker peptides have been evaluated using this technique and demonstrated good inhibition of ThT binding (16, 19, 20, 25, 28, 29). In order to solubilize the hydrophobic peptides, 25% DMSO was required, which was not problematic, as control experiments using 0%, 5%, and 25% DMSO did not show significant differences in the extent of fibril formation. All peptides studied demonstrated some inhibition of fibrillogenesis where somewhat surprisingly the peptides containing one copy of the targeting sequence were generally more effective than the dimers with the exception of compound 16. In general, the peptides showed inhibition numbers that were as good as if not superior to the

1092 Bioconjugate Chem., Vol. 19, No. 5, 2008

Stephenson et al.

Scheme 2. Synthesis of Peptides (11, 12) Containing Two Targeting Sequences and Biotin

values reported for the parent peptides. For example, Soto and co-workers found lvffa had 89 ( 2.7% amyloid when present in a 10-fold excess compared to Aβ1-40, while Zhang et al. prepared the D-analogue of klvff and reported the extent of amyloid formation as 76% (17). The rhenium analogue ReSAACQ-lvffag was more potent and showed 48 ( 4% inhibition when used in a 1:1 ratio with Aβ1-40. These results indicate that the presence of the metal complex and/or biotin derivative did not have a detrimental impact on the ability of the peptide to bind to amyloid. The most promising agents for further evaluation are compounds 8a and 16, both of which contain biotin as a site for further derivatization. The next phase of the work was to demonstrate that the 99mTc analogues could be prepared and that the products are sufficiently robust to screen in ViVo. Compound 7 was selected as a model compound and labeled using [99mTc(CO)3(OH2)3]+, which can be readily prepared in high yield using an instant kit (30-34). The labeled product, k(N
Biotin)-99mTcSAACQlvffag (8b) was isolated by solid-phase extraction and HPLC purification. The latter technique was employed to remove any unlabeled ligand, thus affording the product in high effective

specific activity. Analytical HPLC in fact indicated that the reaction was quantitative; however, the isolated radiochemical yield after semipreparative HPLC was greatly reduced largely due to absorption of the labeled peptide on the column. Nevertheless, the desired product was isolated in high radiochemical purity (greater than 98%), and its retention time matched that of the Re analogue. Ligand challenge studies on the product showed no significant signs of degradation (Figure 4), which is consistent with the d6 electronic configuration of the octahedral metal complex.

CONCLUSIONS An amino acid derived chelating system for 99mTc and Re was used to create a small library of β-breaker peptide derivatives. The Re complexes, which were synthesized in parallel with the free ligands, were tested for their ability to inhibit fibrillogenesis, and two promising leads were identified. The 99mTc analogue of one of these was isolated in modest yield but exquisite purity, and it was resistant to ligand challenge against both histidine and cysteine. Amyloid plaques are associated with a number of diseases beyond simply AD,

Re/Tc Amyloid Probes from β-Breaker Peptides Scheme 3. Labeling of Compound 7 with

99m

Bioconjugate Chem., Vol. 19, No. 5, 2008 1093 Tc

consequently molecular imaging probes for this particular target like those reported here could have significant utility in both clinical applications and in supporting early trials involving emerging therapeutics. Furthermore, the results presented demonstrate that the ReSAACQ system can be used to efficiently generate multiple Re-metallopeptides in parallel, which can be screened directly to identify lead agents in a time and resource efficient manner.

Figure 4. γ-HPLC radiochromatograms of samples from the histidine (a) and cysteine (b) challenge experiments involving 8b (Rt ) 20.6 min) after 24 h at 37 °C.

ACKNOWLEDGMENT We would like to acknowledge funding from the Ontario Research and Development Challenge Fund (ORDCF) and The National Sciences and Engineering Research Council (NSERC) of Canada for a Scholarship for K.A.S. Supporting Information Available: Structures of the ligands; HPLC, IR, and ESMS data. This material is available free of charge via the Internet at http://pubs.acs.org.

LITERATURE CITED

Figure 3. (a) UV-HPLC chromatogram of 8a (Rt ) 19.7 min) and (b) γ-HPLC chromatogram of purified 8b (Rt ) 20.4 min).

(1) Bachurin, S. O. (2002) Medicinal chemistry approaches for the treatment and prevention of Alzheimer’s disease. Med. Res. ReV. 23, 48–88. (2) Mattson, M. P. (2004) Pathways towards and away from Alzheimer’s disease. Nature 430, 631–639. (3) Kurihara, A., and Pardridge, W. M. (2000) Aβ1-40 peptide radiopharmaceuticals for biotin amyloid imaging: 111In chelation, conjugation to poly(ethylene glycol)-biotin linkers, and autoradiography with Alzheimer’s disease brain sections. Bioconjugate Chem. 11, 380–386. (4) Soto, C. (1999) Plaque busters: strategies to inhibit amyloid formation in Alzheimer’s disease. Mol. Med. Today 5, 343–350. (5) Geula, C., Wu, C. K., Saroff, D., Lorenzo, A., Yuan, M., and Yankner, B. A. (1998) Aging renders the brain vulnerable to amyloid β-protein neurotoxicity. Nat. Med. 4, 827–831. (6) Hardy, J., and Selkoe, D. J. (2002) The amyloid hypothesis of Alzheimer’s disease: progress and problems on the road to therapeutics. Science 297, 353–356. (7) Pitschke, M., Prior, R., Haupt, M., and Riesner, D. (1998) Detection of single amyloid β-protein aggregates in the cerebrospinal fluid of Alzheimer’s patients by fluorescence correlation spectroscopy. Nat. Med. 4, 832–834. (8) Verhoeff, N. P., Wilson, A. A., Takeshita, S., Trop, L., Hussey, D., Singh, K., Kung, H. F., Kung, M.-P., and Houle, S. (2004) In vivo imaging of Alzheimer disease β-amyloid with [11C]SB13 PET. Am. J. Geriatr. Psychiatry 12, 584–595.

1094 Bioconjugate Chem., Vol. 19, No. 5, 2008 (9) Klunk, W. E., Engler, H., Nordberg, A., Wang, Y., Blomqvist, G., Holt, D. P., Bergstrom, M., Savitcheva, I., Huang, G.-f., Estrada, S., Ausen, B., Debnath, M. L., Barletta, J., Price, J. C., Sandell, J., Lopresti, B. J., Wall, A., Koivisto, P., Antoni, G., Mathis, C. A., and Langstrom, B. (2004) Imaging brain amyloid in Alzheimer’s disease with Pittsburgh Compound-B. Ann. Neurol. 55, 306–319. (10) Qu, W., Kung, M.-P., Hou, C., Oya, S., and Kung, H. F. (2007) Quick assembly of 1,4-diphenyltriazoles as probes targeting β-amyloid aggregates in Alzheimer’s disease. J. Med. Chem. 50, 3380–3387. (11) Qu, W., Kung, M.-P., Hou, C., Benedum, T. E., and Kung, H. F. (2007) Novel styrylpyridines as probes for SPECT imaging of amyloid plaques. J. Med. Chem. 50, 2157–2165. (12) Chandra, R., Oya, S., Kung, M.-P., Hou, C., Jin, L.-W., and Kung, H. F. (2007) New diphenylacetylenes as probes for positron emission tomographic imaging of amyloid plaques. J. Med. Chem. 50, 2415–2423. (13) Maggio, J. E., Stimson, E. R., Ghilardi, J. R., Allen, C. J., Dahl, C. E., Whitcomb, D. C., Vigna, S. R., Vinters, H. V., Labenski, M. E., and Mantyh, P. W. (1992) Reversible in vitro growth of Alzheimer disease β-amyloid plaques by deposition of labeled amyloid peptide. Proc. Natl. Acad. Sci. U.S.A. 89, 5462–5466. (14) Wu, D., Yang, J., and Pardridge, W. M. (1997) Drug targeting of a peptide radiopharmaceutical through the primate blood-brain barrier in vivo with a monoclonal antibody to the human insulin receptor. J. Clin. InVest. 100, 1804–1812. (15) Saito, Y., Buciak, J., Yang, J., and Pardridge, W. M. (1995) Vector-mediated delivery of 125 I-labeled β-amyloid peptide Aβ1-40 through the blood-brain barrier and binding to Alzheimer disease amyloid of the Aβ1-40/vector complex. Proc. Natl. Acad. Sci. U.S.A. 92, 10227–10231. (16) Chalifour, R. J., McLaughlin, R. W., Lavoie, L., Morissette, C., Tremblay, N., Boule, M., Sarazin, D. S., Lacombe, D., Tremblay, P., and Gervais, F. (2003) Stereoselective interactions of peptide inhibitors with the β-amyloid peptide. J. Biol. Chem. 278, 34874–34881. (17) Zhang, G., Leibowitz, M. J., Sinko, P. J., and Stein, S. (2003) Multiple-peptide conjugates for binding β-amyloid plaques of Alzheimer’s disease. Bioconjugate Chem. 14, 86–92. (18) Gazit, E. (2002) Mechanistic studies of the process of amyloid fibrils formation by the use of peptide fragments and analogues: implications for the design of fibrillization inhibitors. Curr. Med. Chem. 9, 1725–1735. (19) Gordon, D. J., Sciarretta, K. L., and Meredith, S. C. (2001) Inhibition of 102-amyloid(40) fibrillogenesis and disassembly of β-amyloid(40) fibrils by short β-amyloid congeners containing N-methyl amino acids at alternate residues. Biochemistry 40, 8237–8245. (20) Soto, C., Kindy, M. S., Baumann, M., and Frangione, B. (1996) Inhibition of Alzheimer’s amyloidosis by peptides that prevent β-sheet conformation. Biochem. Biophys. Res. Commun. 226, 672–680. (21) Soto, C., Sigurdsson, E. M., Morelli, L., Kumar, R. A., Castano, E. M., and Frangione, B. (1998) β-Sheet breaker peptides inhibit fibrillogenesis in a rat brain model of amyloid-

Stephenson et al. osis: Implications for Alzheimer’s therapy. Nat. Med. 4, 822– 826. (22) Findeis, M. A., Musso, G. M., Arico-Muendel, C. C., Benjamin, H. W., Hundal, A. M., Lee, J.-J., Chin, J., Kelley, M., Wakefield, J., Hayward, N. J., and Molineaux, S. M. (1999) Modified-peptide inhibitors of amyloid β-peptide polymerization. Biochemistry 38, 6791–6800. (23) Chan, W. C., and White, P. D. (2000) FMOC Solid-Phase Peptide Synthesis, A Practical Approach, Oxford University Press, New York. (24) Li, D., and Elbert, D. L. (2002) The kinetics of the removal of the N-methyltrityl (mtt) group during synthesis of branched peptides. J. Pept. Res. 60, 300–303. (25) Poduslo, J. F., Curran, G. L., Kumar, A., Frangione, B., and Soto, C. (1999) Beta-sheet breaker peptide inhibitor of Alzheimer’s amyloidogenesis with increased blood-brain barrier permeability and resistance to proteolytic degradation in plasma. J. Neurobiol. 39, 371–382. (26) Mathis, C. A., Wang, Y., and Klunk, W. E. (2004) Imaging β-amyloid plaques and neurofibrillary tangles in the aging human brain. Curr. Pharm. Des. 10, 1469–1492. (27) LeVine, H. I. (1993) Thioflavine T interaction with synthetic Alzheimer’s disease β-amyloid peptides: Detection of amyloid aggregation in solution. Protein Sci. 2, 404–410. (28) Adessi, C., Frossard, M.-J., Boissard, C., Fraga, S., Bieler, S., Ruckle, T., Vilbois, F., Robinson, S. M., Mutter, M., Banks, W. A., and Soto, C. (2003) Pharmacological profiles of peptide drug candidates for the treatment of Alzheimer’s disease. J. Biol. Chem. 278, 13905–13911. (29) Hetenyi, C., Kortvelyesi, T., and Penke, B. (2002) Mapping of possible binding sequences of two β-sheet breaker peptides on β-amyloid peptide of Alzheimer’s disease. Bioorg. Med. Chem. 10, 1587–1593. (30) Egli, A., Alberto, R., Tannahill, L., Schibli, R., Abram, U., Schaffland, A., Waibel, R., Tourwe, D., Jeannin, L., Iterbeke, K., and Schubiger, P. A. (1999) Organometallic 99mTc-aquaion labels peptide to an unprecedented high specific activity. J. Nucl. Med. 40, 1913–1917. (31) Alberto, R., Schibli, R., Egli, A., Schubiger, A. P., Abram, U., and Kaden, T. A. (1998) A novel organometallic aqua complex of technetium for the labeling of biomolecules: synthesis of [99mTc(OH2)3(CO)3]+ from [99mTcO4]- in aqueous solution and its reaction with a bifunctional ligand. J. Am. Chem. Soc. 120, 7987–7988. (32) Alberto, R., Schibli, R., Waibel, R., Abram, U., and Schubiger, A. P. (1999) Basic aqueous chemistry of [M(OH2)3(CO)3]+ (M)Re, Tc) directed towards radiopharmaceutical application. Coord. Chem. ReV. 190-192, 901–919. (33) Alberto, R., Ortner, K., Wheatley, N., Schibli, R., and Schubiger, A. P. (2001) Synthesis and properties of boranocarbonate: a convenient in situ CO source for the aqueous preparation of [99mTc(OH2)3(CO)3]+. J. Am. Chem. Soc. 123, 3135–3136. (34) Jaouen, G., Top, S., Vessieres, A., and Alberto, R. (2000) New paradigms for synthetic pathways inspired by bioorganometallic chemistry. J. Organomet. Chem. 600, 23–26. BC800001G