Frank J. Femia

Preclinical Evaluation of Novel Glutamate-Urea-Lysine Analogues That Target Prostate-Specific Membrane Antigen as Molecular Imaging Pharmaceuticals for Prostate Cancer

Prostate-specific membrane antigen (PSMA) is expressed in normal human prostate epithelium and is highly up-regulated in prostate cancer. We previously reported a series of novel small molecule inhibitors targeting PSMA. Two compounds, MIP-1072, (S)-2-(3-((S)-1-carboxy-5-(4-iodobenzylamino)pentyl)ureido)pentanedioic acid, and MIP-1095, (S)-2-(3-((S)-1carboxy-5-(3-(4-iodophenyl)ureido)pentyl)ureido)pentanedioic acid, were selected for further evaluation. MIP-1072 and MIP-1095 potently inhibited the glutamate carboxypeptidase activity of PSMA (K(i) = 4.6 +/- 1.6 nmol/L and 0.24 +/- 0.14 nmol/L, respectively) and, when radiolabeled with (123)I, exhibited high affinity for PSMA on human prostate cancer LNCaP cells (K(d) = 3.8 +/- 1.3 nmol/L and 0.81 +/- 0.39 nmol/L, respectively). The association of [(123)I]MIP-1072 and [(123)I]MIP-1095 with PSMA was specific; there was no binding to human prostate cancer PC3 cells, which lack PSMA, and binding was abolished by coincubation with a structurally unrelated NAALADase inhibitor, 2-(phosphonomethyl)pentanedioic acid (PMPA). [(123)I]MIP-1072 and [(123)I]MIP-1095 internalized into LNCaP cells at 37 degrees C. Tissue distribution studies in mice showed 17.3 +/- 6.3% (at 1 hour) and 34.3 +/- 12.7% (at 4 hours) injected dose per gram of LNCaP xenograft tissue, for [(123)I]MIP-1072 and [(123)I]MIP-1095, respectively. [(123)I]MIP-1095 exhibited greater tumor uptake but slower washout from blood and nontarget tissues compared with [(123)I]MIP-1072. Specific binding to PSMA in vivo was shown by competition with PMPA in LNCaP xenografts, and the absence of uptake in PC3 xenografts. The uptake of [(123)I]MIP-1072 and [(123)I]MIP-1095 in tumor-bearing mice was corroborated by single-photon emission computed tomography/computed tomography (SPECT/CT) imaging. PSMA-specific radiopharmaceuticals should provide a novel molecular targeting option for the detection and staging of prostate cancer.

Novel Polar Single Amino Acid Chelates for Technetium-99m Tricarbonyl-Based Radiopharmaceuticals with Enhanced Renal Clearance: Application to Octreotide

Single amino acid chelate (SAAC) systems for the incorporation of the M(CO)(3) moiety (M = Tc/Re) have been successfully incorporated into novel synthetic strategies for radiopharmaceuticals and evaluated in a variety of biological applications. However, the lipophilicity of the first generation Tc(CO)(3)-dipyridyl complexes has resulted in substantial hepatobiliary uptake when either examined as lysine derivatives or integrated into biologically active small molecules and peptides. Here we designed, synthesized, and evaluated novel SAAC systems that have been chemically modified to promote overall Tc(CO)(3)L(3) complex hydrophilicity with the intent of enhancing renal clearance. A series of lysine derived SAAC systems containing functionalized polar imidazole rings and/or carboxylic acids were synthesized via reductive alkylation of the epsilon amino group of lysine. The SAAC systems were radiolabeled with (99m)Tc, purified, and evaluated for radiochemical stability, lipophilicity, and tissue distribution in rats. The log P values of the (99m)Tc complexes were determined experimentally and ranged from -0.91 to -2.33. The resulting complexes were stable (>90%) for at least 24 h. Tissue distribution in normal rats of the lead (99m)Tc complexes demonstrated decreased liver (<1 %ID/g) and gastrointestinal clearance (<1.5%ID/g) and increased kidney clearance (>15 %ID/g) at 2 h after injection compared to the dipyridyl lysine complex (DpK). One of the new SAAC ligands, [(99m)Tc]bis-carboxymethylimidazole lysine, was conjugated to the N-terminus of Tyr-3 octreotide and evaluated for localization in nude mice bearing AR42J xenografts to examine tissue distribution, tumor uptake and retention, clearance, and route of excretion for comparison to (111)In-DOTA-Tyr-3-octreotide and (99m)Tc-DpK-Tyr-3-octreotide. (99m)Tc-bis-(carboxymethylimidazole)-lysine-Tyr-3-octreotide exhibited significantly less liver uptake and gastrointestinal clearance compared to (99m)Tc-DpK-Tyr-3-octreotide while maintaining tumor uptake in the same mouse model. These novel chelators demonstrate that lipophilicity can be controlled and organ distribution significantly altered, opening up broad application of these novel SAAC systems for radiopharmaceutical design.

Comprehensive Radiolabeling, Stability, and Tissue Distribution Studies of Technetium-99m Single Amino Acid Chelates (SAAC)

Technetium tricarbonyl chemistry has been a subject of interest in radiopharmaceutical development over the past decade. Despite the extensive work done on developing chelates for Tc(I), a rigorous investigation of the impact of changing donor groups and labeling conditions on radiochemical yields and/or distribution has been lacking. This information is crucially important if these platforms are going to be used to develop molecular imaging probes. Previous studies on the coordination chemistry of the {M(CO)(3)}(+) core have established alkylamine, aromatic nitrogen heterocycles, and carboxylate donors as effective chelating ligands. These observations led to the design of tridentate ligands derived from the amino acid lysine. Such amino acid analogues provide a tridentate donor set for chelation to the metal and an amino acid functionality for conjugation to biomolecules. We recently developed a family of single amino acid chelates (SAAC) that serve this function and can be readily incorporated into peptides via solid-phase synthesis techniques. As part of these continuing studies, we report here on the radiolabeling with technetium-99m ((99m)Tc) and stability of a series of SAAC analogues of lysine. The complexes studied include cationic, neutral, and anionic complexes. The results of tissue distribution studies with these novel complexes in normal rats demonstrate a range of distribution in kidney, liver, and intestines.

Comparison of High-Specific-Activity Ultratrace 123/131I-MIBG and Carrier-Added 123/131I-MIBG on Efficacy, Pharmacokinetics, and Tissue Distribution

Metaiodobenzylguanidine (MIBG) is an enzymatically stable synthetic analog of norepinephrine that when radiolabled with diagnostic ((123)I) or therapeutic ((131)I) isotopes has been shown to concentrate highly in sympathetically innervated tissues such as the heart and neuroendocrine tumors that possesses high levels of norepinephrine transporter (NET). As the transport of MIBG by NET is a saturable event, the specific activity of the preparation may have dramatic effects on both the efficacy and safety of the radiodiagnostic/radiotherapeutic. Using a solid labeling approach (Ultratrace), noncarrier-added radiolabeled MIBG can be efficiently produced. In this study, specific activities of >1200 mCi/micromol for (123)I and >1600 mCi/micromol for (131)I have been achieved. A series of studies were performed to assess the impact of cold carrier MIBG on the tissue distribution of (123/131)I-MIBG in the conscious rat and on cardiovascular parameters in the conscious instrumented dog. The present series of studies demonstrated that the carrier-free Ultratrace MIBG radiolabeled with either (123)I or (131)I exhibited similar tissue distribution to the carrier-added radiolabeled MIBG in all nontarget tissues. In tissues that express NETs, the higher the specific activity of the preparation the greater will be the radiopharmaceutical uptake. This was reflected by greater efficacy in the mouse neuroblastoma SK-N-BE(2c) xenograft model and less appreciable cardiovascular side-effects in dogs when the high-specific-activity radiopharmaceutical was used. The increased uptake and retention of Ultratrace (123/131)I-MIBG may translate into a superior diagnostic and therapeutic potential. Lastly, care must be taken when administering therapeutic doses of the current carrier-added (131)I-MIBG because of its potential to cause adverse cardiovascular side-effects, nausea, and vomiting.

Cationic complexes of the '3 + 1' oxorhenium-thiolate family

Abstract The ‘3+1’ concept of ligand addition to Group VII metaloxo species allows the facile preparation of neutral complexes of the general type [MO(SXY)(SR)], where X=S, O, or N and Y=S or O. However, if appropriate amine functionalities are incorporated into the substituent R of the monodentate ligand, cationic oxometalate complexes of the type [MO(SXY)(SRNH x )] + may be routinely prepared. The synthesis and characterization of a series of complexes of this general class are reported. The structures of [ReO{(SCH 2 CH 2 ) 2 S}(C 5 H 4 NH-2-S)][Br], [ReO{(SCH 2 CH 2 ) 2 S}(C 5 H 4 NH-4-S)][Br], [ReO{(SCH 2 CH 2 ) 2 S}(SCH 2 CH 2 NH 3 )][Cl], and [ReO{C 5 H 3 N-2-CH 2 O-6-CH 2 S}(C 5 H 4 NH-2-S)}][Br] are described and compared to other examples of ‘3+1’ oxorhenium thiolate complexes.

Synthesis and structural characterization of rhenium(I) tricarbonyl complexes with the bidentate ligands o-(diphenylphosphino)benzaldehyde (P∩O) and o-[(diphenylphosphino)benzylidene]analine (P∩N).

A series of rhenium(I) tricarbonyl complexes with bidentate P,O donor ligand o-(diphenylphosphino)benzaldehyde (P∩O) and its Schiff base P,N donor ligand o-[diphenylphosphino)benzylidene]analine (P∩N) have been synthesized and structurally characterized. All the complexes of the type [ReX(CO)(3)(LL)] (where LL = P∩O, P∩N) reveal a distorted octahedral structure with the three carbonyl ligands arranged in the facial fashion. Crystal data for 1, C(22)H(15)ClO(4)PRe·1/2C(6)H(14): triclinic, P1, a=8.7430(4), b=9.5767(4), c=13.9449(6) Å, α=93.651(1), β=101.265(1), γ=93.048(1); V=1140.20(9) Å(3), Z=2.2, C(22)H(15)BrO(4)Pre: monoclinic, C2/c, a=31.6800(16), b=8.8880(4), c=18.4517(9) Å, β=124.990(1); V=4256.4(4) Å(3), Z=8. 3, C(28)H(20)ClNO(3)Pre: monoclinic, C2/c, a=22.675(4), b=8.803(2), c=28.218(5) Å, β=100.192(3); V=5543.5(16) Å(3), Z=8.4, C(28)H(20)BrNO(3)Pre: monoclinic, C2/c, a=23.035(1), b=8.7561(4), c=28.269(1) Å, β=100.811(1); V=5600.4(5) Å(3), Z=8.

Molecular imaging of human ACE-1 expression in transgenic rats.

OBJECTIVES The aim of this study was to develop a molecular imaging strategy that can monitor myocardial angiotensin-converting enzyme (ACE)-1 upregulation as a function of progressive heart failure. BACKGROUND High-affinity technetium-99m-labeled lisinopril (Tc-Lis) has been shown to specifically localize in tissues that express ACE in vivo, such as the lungs. Whether Tc-Lis can also detect upregulation of ACE in the heart, by external in vivo imaging, has not been established. METHODS Twenty-one ACE-1 over-expressing transgenic (Tg) and 18 wild-type control rats were imaged using in vivo micro single-positron emission computed tomography (SPECT)-computed tomography (CT) at 10, 30, 60, and 120 min after Tc-Lis injection. A subgroup of rats received nonradiolabeled (cold) lisinopril before the Tc-Lis injection to evaluate nonspecific binding. After imaging, the rat myocardium was explanted, ex vivo images were acquired, and percent injected dose per gram gamma-well was counted, followed by an assessment of enzyme-linked immunosorbent assay-verified ACE activity and messenger ribonucleic acid expression. RESULTS On micro SPECT-CT, myocardial ACE-1 uptake was best visualized in Tg rats at 120 min after Tc-Lis injection. The quantitative uptake of Tc-Lis in the myocardium was 5-fold higher in mutant Tg than in control rats at each time point after tracer injection. The percent injected dose per gram uptake was 0.74 ± 0.13 in Tg myocardium at 30 min and was reduced substantially to 0.034 ± 0.003% when pre-treated with cold lisinopril (p = 0.029). Enzyme activity assay showed a >30-fold higher level of ACE-1 activity in the myocardium of Tg rats than in controls. The ACE-1 messenger ribonucleic acid was quantified, and lisinopril was found to have no effect on ACE-1 gene expression. CONCLUSIONS The Tc-Lis binds specifically to ACE, and the activity can be localized in Tg rat hearts that over-express human ACE-1 with a signal intensity that is sufficiently high to allow external imaging. Such a molecular imaging strategy may help identify susceptibility to heart failure and may allow optimization of pharmacologic intervention.

Synthesis and Evaluation of a Series of 99mTc(CO)3+ Lisinopril Complexes for In Vivo Imaging of Angiotensin-Converting Enzyme Expression

In animal models of cardiac disease and in human congestive heart failure, expression of angiotensin-converting enzyme (ACE) is upregulated in the failing heart and has been associated with disease progression leading to cardiac failure and fibrosis. To develop probes for imaging ACE expression, a series of di(2-pyridylmethyl)amine (D) chelates capable of binding M(CO)3+ (M = technetium, rhenium) was conjugated to lisinopril by acylation of the ε-amine of the lysine residue with a series of di(2-pyridylmethylamino)alkanoic acids where the distance of the chelator from the lisinopril core was investigated by varying the number of methylene spacer groups to produce di(2-pyridylmethyl)amine(Cx)lisinopril analogs: D(C4)lisinopril, D(C5)lisinopril, and D(C8)lisinopril. The inhibitory activity of each rhenium complex was evaluated in vitro against purified rabbit lung ACE and was shown to vary directly with the length of the methylene spacer: Re[D(C8)lisinopril], inhibitory concentration of 50% (IC50) = 3 nM; Re[D(C5)lisinopril], IC50 = 144 nM; and Re[D(C4)lisinopril], IC50 = 1,146 nM, as compared with lisinopril, IC50 = 4 nM. The in vivo specificity for ACE was determined by examining the biodistribution of the 99mTc-[D(C8)lisinopril] analog in rats with and without pretreatment with unlabeled lisinopril. Uptake in the lungs, a tissue that constitutively expresses ACE, was 15.2 percentage injected dose per gram at 10 min after injection and was dramatically reduced by pretreatment with lisinopril, supporting ACE-mediated binding in vivo. Planar anterior imaging analysis of 99mTc-[D(C8)lisinopril] corroborated these data. Thus, high-affinity 99mTc-labeled ACE inhibitor has been designed with potency similar to that of lisinopril and has been demonstrated to specifically localize to tissues that express ACE in vivo. This agent may be useful in monitoring ACE as a function of disease progression in relevant diseases such as heart failure.

Expansion of the ‘3 + 1’ concept of oxorhenium-thiolate chemistry to cationic and binuclear complexes

Abstract The reactions of [(n-C4H9)4N] [ReOBr4(OPPh3)] with the appropriate tridentate and monodentate thiolate ligands yield the ‘3 + 1’ oxorhenium(V) complexes [ReO(SCH2CH2OCH2CH2S)(SCH2C6H4-4-OMe)] (1), [ReO(SCH2CH2OCH2CH2S)(SC5H4NH)]Br (2) and [Re2O2(SCH2CH2OCH2CH2S)3] (3). The versatility of this approach for the synthesis of oxorhenium(V) core complexes is manifest in the preparations of the cationic complex of 2 and of the thiolate bridged binuclear complex 3.

Synthesis, characterization, and biodistribution of a Technetium-99m ‘3+1’ fatty acid derivative. The crystal and molecular structures of a series of oxorhenium model complexes

Abstract The ‘3+1’ design of complexes of the {ReVO}3+ core was exploited in the preparation of two series of compounds incorporating fatty acid components. In one case, the fatty acid subunit is attached to the monodentate ligand ‘1’ of the [ReO(tridentate)(monodentate)] complex, with the tridentate component consisting of the sulfido–dithiolate donor, {(SCH2CH2)2S}2−, and represented by the complexes [ReO{η3-(SCH2CH2)2S}{η1-S(CH2)4CO2H}] (1) and [ReO{η3-(SCH2CH2)2S}{η1-S(CH2)10CO2H}] (2). The second series exhibits the fatty acid attachment at the tridentate site, {(SCH2CH2)2NR}2−, in the complexes of the general type [ReO{η3-(SCH2CH2)2NR}{η1-SC6H4X}] (4: R=–(CH2)7CH3, X=–Cl; 5: R=–(CH2)7CH3, X=–Br; 6: R=–(CH2)7CH3, X=–OCH3) and [ReO{η3-(SCH2CH2)2NR}{η1-SCH2C6H4X}] (7: R=–(CH2)7CH3, X=Cl; 8: R=–(CH2)15CH3, X=–H). Compounds 1 and 2 exhibit square-pyramidal geometry, while 4–8 approach more closely the trigonal bipyramidal prototype. The 99mTc analogue [99mTcO{η3-(SCH2CH2)2N(CH2)15CO2H}{η1-SCH2C6H5}] (9) was also prepared. Preliminary biodistribution studies in rats indicate poor heart-to-blood ratios (less than 1.0), suggesting that the complex is not a useful reagent for heart imaging.