ORIGINAL ARTICLE 
Annals of Nuclear Medicine Vol. 10, No. 1, 19-32, 1996 

Both total chain length and position of dimethyl-branching effect the myocardial uptake and retention of radioiodinated analogues of 15-(p-iodophenyl)-3,3-dimethylpentadecanoic acid (DMIPP) 

F. F. KNAPP Jr, * M. M. GOODMAN, *,1 G. KIRSCH, *,2 S. N. RESKE,**,3 J. KRopp,**,4 H.-J. BIERSACK,** K.R. AMBROSE,* C.R. LAMBERT* and A. GOUDONNET* 

*Nuclear Medicine Group. Health Sciences Research Division, Oak Ridge National Laboratory (ORNL), USA 
**Clinicfor Nuclear Medicine, University of Bonn, Germany 
1Department of Radiology, Emory University Hospital, Atlanta, Georgia 
2Department of Chemistry, University of Metz, Metz, France 
3Department of Nuclear Medicine, University Hospital, Ulm. Germany 
4Clinic and Polyclinic for Nuclear Medicine. Technical University, Dresden, Germany 

Introduction of geminal dimethyl-branching into the 3-position of 15-(p-iodophenyl)pentadecanoic acid (IPPA) significantly delays myocardial clearance in rats and dogs following intravenous administration. Several new analogues of DMIPP have been synthesized and evaluated in fasted rats. The effects of both the position of dimethyl-branching and the total chain-length of 3,3-dimethyl analogues on heart uptake and clearance kinetics have been studied. In the first series of compounds, two methyl groups were introduced into the 3-, 4-, 6-, or 9-position. Tissue distribution studies of the 15-(p-[I-125]iodophenyl)-analogues demonstrated that the position of dimethyl-branching is an important factor affecting both myocardial specificity and retention. The [I-125]labeled 3,3- and 4,4-DMIPP analogues showed higher myocardial uptake and faster blood clearance than the 6,6- and 9,9-DMIPP analogues [heart, % dose/gm (heart : blood), 30 min: 3,3-DMIPP = 5.06 (12 : 1); 4,4-DMIPP = 8.03 (16.7 : 1); 6,6-DMIPP = 2.26 (3.1 : 1); 9,9-DMIPP = 3.06 (2.77)]. In the second series, the effects of total fatty acid chain length were evaluated with 3,3-dimethyl-substituted analogues with C11, C12, C13, C14, C15, and C19 chain lengths. The C14 and C15 chain length analogues showed the best properties [global heart uptake (heart : blood ratios): 30 min: C11, 0.70 (0.82); C12, 1.25 (0.68); C13, 0.47 (0.90); C14, 1.63 (3.54); C15, 5.06 (12); C19 1.29 (0.82). These detailed studies have demonstrated that both total chain length and the position of geminal dimethyl-branching are important structural parameters which affect myocardial specificity and retention of W>(p-iodophenyl)-substituted fatty acid analogues and that 3,3-DMIPP and 4,4-DMIPP are the best candidates with optimal properties for further study. 

Key words: cardiac SPECT, cardiodine, DMIPP, iodine-123, methyl-branched fatty acids 

INTRODUCTION 
THE AVAILABILITY of iodine-123 labeled fatty acid analogues for the evaluation of myocardial fatty acid energy substrate uptake and metabolism with single photon emission computed tomography (SPECT) has continued to stimulate widespread interest.1-6 Our development of 15-(p-iodophenyl)-3-R,S-methylpentadecanoic acid (BMIPP; Fig. 1; R,S-notation indicates a racemic mixture) involved introduction of a methyl group into the 3-position of the 15-(p-iodophenyl)pentadecanoic acid (IPPA). The iodine-123-BMIPP is now commercially available as "Cardiodine" from Nihon Medi-Physics, Inc., an approved cardiac imaging agent. Radioiodinated BMIPP shows excellent myocardial localization and considerably longer retention in rats7,8 and dogs9 and this agent has proven to be an important tool in cardiac research. Examples include the use of radioiodinated BMIPP for the evaluation of aberrations in regional fatty acid uptake in rat hearts using autoradiography,(10,11) evaluation of uptake and washout in an occlusion-reperfusion canine model(12) and investigation of the effects of high energy phosphate (ATP) levels on myocafdial fatty acid uptake in mice using the 2,4-dinitrophenol (DNP) uncoupler. (13,14) BMIPP analogues have also been used by several groups for the evaluation of fatty acid uptake and metabolism by both working and non-working Langendorff isolated rat heart models.(15-18) Subsequent to our studies, an analogue of BMIPP in which the racemic methyl-substitution has been introduced into the 9-position has been synthesized(l9) and evaluated in an infarction model in dogs(20) and more recently in patients with exercise-induced ischemia.(21) 
 Clinical use of iodine-123-labeled "modified" fatty acids analogues which show delayed myocardial clearance for myocardial single photon computerized tomographic (SPECT) imaging results from differences between regional myocardial fatty acid uptake patterns and flow tracer distribution which are often observed in various myocardial disorders. (1-6) The use of Cardiodine in conjunction with flow tracers may represent a unique opportunity to correlate energy substrate metabolism with regional myocardial viability using SPECT. Although the physiological basis is not completely understood, differences between regional fatty acid and flow tracer distribution may reflect alterations in important parameters of metabolism which can be useful for planning patient management and therapy. 
The role of Cardiodine in nuclear cardiology is rapidly being defined and a variety of clinical studies have shown the usefulness of this agent with dual radioisotope studies in conjunction with flow tracers for detection and characterization of patients with hypertrophic cardiomyopathy.(22-25) The data clearly demonstrate that reduced regional accumulation of BMIPP in comparison with thallium is often observed in hypertrophic myocardium and that Cardiodine SPECT in conjunction with flow tracer assessment is thus a useful new technique to assess functional integrity of the myocardium. In spite of the fact that the physiological factors affecting such distribution patterns is not well understood the significant differences often observed between flow tracer distribution and BMIPP uptake are felt to represent abnormalities of myocardial fatty acid metabolism in cardiomyopathic myocardium, reflecting an "intrinsic" impairment of myocardial free fatty acid utilization. 
 In patients with myocardial infarction and ischemia, the regional uptake and clearance of Cardiodine has also been evaluated in a number of studies.(22,26-28) "Mismatch" between BMIPP uptake and perfusion tracer distribution (i.e. Thallium-201 > BMIPP) is often observed in areas which had suffered an acute myocardial infarction and areas which were supplied by revascularized compared with nonrevascularized areas. The regional myocardial distribution of Cardiodine at rest has also been compared with [Tc-99m]-Sestamibi (MIBI).(29) The important finding from these combined studies is that lower BMIPP uptake is consistently more often observed in segments which exhibited wall motion scores lower than perfusion scores in comparison to segments showing a concordant decrease in both wall motion and perfusion scores.(29-31) 
 Most importantly, Cardiodine SPECT may represent a new clinical tool which could be used routinely where PET is not available for the differentiation of viable myocardium from scar following thrombolysis for acute myocardial infarction. In this regard, comparison of Cardiodine uptake, ventricular function and [18F]-labeled-2-FDG has evaluated the factors which result in deceased fatty extraction in myocardial segments that have adequate perfusion but demonstrate significantly decreased contractile function. The regional uptake and clearance kinetics of Cardiodine and [1-ll-C]palmitate with successive SPECT and PET studies have been recently evaluated in the same patients by Tamaki, et al.32 Comparison of PET FDG results with Cardiodine SPECT clearly suggests that Cardiodine/thallium-201 mis-match is significant and can be corroborated by PET data.(32,33)
 Based on the expected interference of the C-3 monomethyl group on subsequent catabolism of the fatty acid chain, we envisaged that introduction of two methyl groups (i.e. dimethyl-substitution) at the 3-position may impart complete inhibition of carbon-carbon bond cleavage and result in essentially irreversible myocardial retention.(34-36) Initial studies with the 3,3-dimethyl-branchedanalogue, 1 5-(p-iodophenyl)-3,3-dimethyl-pentadecanoic acid ("DMIPP" the dimethyl analogue of "Cardiodine"; Fig. 1), have demonstrated rapid and pronounced myocardial uptake in rats and dogs with essentially irreversible retention over 60 min following i.v. administration.(34,37) 
 The goals of the present study were to further evaluate the important effects of dimethyl-substitution by synthesizing and systematically testing myocardial uptake and retention in rats several new analogues in which we varied both the position of dimethyl-branching and the total chain length. Nine new iodine-125-labeled dimethyl-substituted terminal p-iodophenyl fatty acid analogues have been synthesized and evaluated. The four analogues in the first series have the same 15-carbon aliphatic chain length as Cardiodine with dimethyl-branching introduced at the C-3, C-4, C-6, and C-9 positions. The second series consists of five 3,3-dimethyl-branched analogues with total carbon chain lengths of Cll, C12, C11, C14, C15 (dimethyl analogue of Cardiodine) and C19. The availability of these nine analogues has now permitted the first systematic evaluation of the effects of both chain length and dimethyl substitution on myocardial uptake, retention, and heart : blood ratios. A preliminary report of the effects of the position of dimethyl-branching in IPPA ana-logues on heart uptake in rats has been discussed.(36) 

MATERIALS AND METHODS 

General

All solvents, chemicals and reagents were analytical grade and were used without further purification unless otherwise indicated. The melting points were determined on a Thomas-Hoover apparatus in open capillary tubes and are uncorrected. The proton nuclear magnetic resonance spectra (NMR) was determined at 60 MHz using a Varian 360-A insrument in the solvents indicated and resonances (ppm) are reported as downfield (d) from an internal tetramethylsilane standard. The thin-layer chromatographic (TLC) analyses were performed using precoated 250 nm layers of SiO2 glass plates (Analtech, Inc.). The mass spectral analyses (MS) were determined by the Organic Spectroscopy Group, Analytical Chemistry Division at ORNL, by laser ionization FT mass spectrometry. All final compounds [9,24,28e-28f] were chromatographically pure and exhibited mass spectral and nuclear magnetic resonance (lH) spectral properties which were consistent with the assigned structures.

Friedal-Crafts Acylation of Thiophenes-General Procedure 

This procedure was used for the acylation reactions as described earlier.(7) For acylation of thiophenes containing carboxyl substituent, the carboxyl was converted to the methyl ester by initial conversion to the acid chloride by refluxing with a 10% excess of SOCl2, and the resulting acyl chloride was then slowly added to MeOH. After evaporation of the MeOH, the product was dissolved in ether and filtered through a short SiO2 column and the solvent evaporated to provide the substrate for the second acylation step. The thiophene substrate was combined with a lO% excess of the acid chloride acylating agent in 1OO mL of CH2Cl2 at Odeg.C. A two molar excess of SnCl2 was added dropwise and the solution then warmed to room temperature and allowed to stir overnight. Excess 6 M HCl was slowly added and the organic layer washed thoroughly with lO% HCl, H20, lO% NaOH and then dried over anhydrous Na2SO4 or MgSO4. Evaporation of the CH2Cl2 then provided the substituted thiophene product. 

Wolff-Kishner Reduction of Thiophene Keto Esters-General Procedure 

The initial part of this procedure consisted of conversion of the acid to the methyl ester by refluxing with excess SOCl2 and subsequent treatment with MeOH as described above. For reduction, the ketone was refluxed with 30% excess of 85% hydrazine hydrate and a ten-fold excess of KOH in 100 ml diethylene glycol for 1-1.5 h. Water was removed by distillation and the mixture then refluxed 3-4 h. The reaction was cooled to room temperature, diluted with a large excess of water and acidified with 1 M HCl to pH 2-3. The product was extracted with ether and the organic layer washed thoroughly with H2O, dried and the solvent removed in vacuo to provide the deoxygenated product. 

Preparation of (p-Iodophenyl)dimethyl-substituted Fatty Acids-General Procedure 

 The acid (0.5 mmol) and thallium(III)trifiuoroacetate (405 mg, 0.75 mmol) in 3 ml of trifiuoroacetic acid were stirred at room temperature under red lights for 5 d. Potassium iodide (415 mg, 2.5 mmol) in 2 ml of H2O was then added and the mixture stirred for 15 min. Sodium thiosulfate (0.5 g) was then added and the mixture stirred an additional 15 min, poured into 50 mL of H2O, and extracted several times with Et20. The combined ether extracts were thoroughly washed with H20 and dried over anhydrous Na2SO4, and the solvent removed in vacuo. The crude product was applied to a silicic acid (30 g) column slurried in C6H6. Fractions (20 ml) 1-lO were eluted with C6H6 followed by elution with 20 mL fractions of CH2C12. Fractions 13-20 were combined and the solvent was removed in vacuo to afford iodinated products 9, 18, and 24 (Fig. 2), and 28a-28f (See Fig. 3). The chemical purity was confirmed by TLC (SiO2-GF) in CHCl3-CH30H 96 : 4, Rf = 0.5. 

Synthesis of 15-(p-iodophenyl)-4,4-dimethylpentade-canoic Acid (4,4-DMIPP: Scheme 1) 2 -(3,3-Dimethyl-5-hydroxypentyl)thiophene (2). The acid 1 (1.84 g, 8.5 mmol) was added dropwise over a 15min period to a stirred suspension of LiAlH4 (1.5 g, 0.04 Mol) in 50 mL of Et2O under an argon atmosphere at room temperature. The resulting mixture was heated under reflux for 4 h and cooled. The unreacted LiAlH4 was decomposed by a dropwise addition of ethyl acetate (5ml) 
and H2O (5ml). The resulting reaction mixture was carefully poured into ice-H2O (1OO ml), acidified with 1O% H2SO4, and extracted thoroughly with Et2O. The combined Et2O extracts were washed thoroughly with H2O and dried over anhydrous Na2SO4, and the solvent was removed in vacuo. The crude product was purified on a column of silica gel (30g). The product was eluted with benzene to give 1.4 g (85%) of 2 as a light yellow oil; NMR (CDCl3) D 1.05 (s, 6H, (CH3)2), 1.7 (m, 2H, thienyl-CH2-CH2-), 1.7 (m, CH2OH-CH2-), 2.85 (m, 2H, thienyl-CH2), 3.5 (t, 3 = 6 Hz, 2H, CH2OH), 6.9-7.25 (m, 3H, aromatic). 
 2- (3, 3-Dimethyl-5-chloropentyl)thiophene (3). Thionyl chloride (3.6 g, 0.03 mol) in 10 ml of CHCl3 was added dropwise over a 30 min period to a solution of 2 (3.9 g, 0.02 mol) and dimethylaniline (3.6 g, 0.02 mol) in 10 ml of CHCl3 vigorously stirred at 0-5C. The solution was stirred an additional 10 min at 0-5C and then heated on a steam bath for 60 min and poured into 100 ml of ice-cold 1M HCl. The aqueous phase was separated and extracted thoroughly with CHCl3. The CHCl3 fractions were combined and washed with 1 M HCl, 5% NaHCO3, and H2O and dried over anhydrous MgSO4, and CHCl3 was removed in vacuo. The crude material was dissolved in petroleum ether and applied to a column containing silica gel slurried in petroleum ether (100 g). Fractions (20 ml in volume) were eluted with petroleum ether (1-1O) followed by fractions (20 ml in volume) with petroleum ether-benzene 3 : 1 (11-25). Fractions 8-19 were combined to give 3 (2.34 g, 53%) as a light yellow oil; NMR (CDCl3) D 1.25 (s, 6H, (CH3)2), 1.7 (m, 2H, thienyl-CH2-CH2-), 1.7 (m, CH2Cl-CH2-), 2.85 (m, 2H, thienyl-CH2), 3.5 (t, J = 7 Hz, 2H, CH2Cl), 6.9-7.25 (m, 3H, aromatic). 
 2-(3,3-Dimethyl-5-chloropentyl)-5-(l-oxo-5-phenylpentyl)thiophene (5). The acid chloride 4 (1.5 g, 7.6 mmol) was added to a solution of thiophene 3 (1.65 g, 7.6 mmol) in 50 ml of CH2Cl2. The resulting mixture was cooled to O'C and anhydrous SnCl4 (2.08 g, 8 mmol) was added dropwise. The solution was stirred at 0C for 30 min then allowed to warm to room temperature and stirred an additional 2 h. The resulting purple-colored solution was then treated with 6 M HCl until an amber solution was obtained. The CH2Cl2 layer was separated and thoroughly washed stepwise with 1O% HCl, H2O and 1O% NaOH, and dried over anhydrous MgSO4. Evaporation of the CH2Cl2 in vacuo yielded 2.2 g (77%) of 5 as an oil; NMR (CDCl2) D 1.05 (s, 6H, (CH3)2), 1.5-1.9 (m, 1OH, CH2), 2.5-3.0 (m, 6H, thienyl-CH2-, and Ph-CH2-), and thienyl-C = O-CH2-), 3.5 (t, J = 7 Hz, 2H, CH2Cl), 6.9 (d, 1H, J = 4 Hz, aromatic), 7.27 (s, 5H, aromatic), 7.53 (d, 1H, J = 4 Hz, aromatic). 
 2-(3,3-Dimethyl-5-cyanopentyl)-5-(l-oxo-5-phenyl-pentyl)thiophene (6). A mixture of 5 (2.0 g, 5.3 mmol), Nal (0.15 g, 1 mmol), and sodium cyanide (0.4 g, 7.95 
mmol) was stirred at 60C for 6 h in 25 ml of dimethyl sulfoxide (DMSO). The mixture was cooled to room temperature poured into 250 ml of H2O and extracted several times with CH2C12. The combined CH2Cl2 extracts were thoroughly washed with H20 and dried over anhydrous MgSO4 and the solvent removed in vacuo. The crude product was dissolved in benzene and applied to a column containing silica gel (30 g) slurried in benzene. Fractions (5 ml in volume) were eluted with benzene. Fractions 16-24 were combined to give 0.66 g of starting material, 5, and fractions 26-4O were combined to give 2(3,3-dimethyl-5-cyanopentyl)-5-(1-oxo-5-phenyl-pentyl)thiophene (6;0.80g,40%) as a colorless oil; NMR (CDC13) d 1.05 (s, 6H, (CH3)2, 1.5-1.9 (m, lOH, CH2); 2.21 (d, 2H, J = 5 Hz, CH2C = N), 2.5-3.0 (m 6H thienyl-CH2-,Ph-CH2-,thienyl-C=O-CH2-), 6.9 (d, 1H, J = 4 Hz, aromatic), 7.27 (s, 5H, aromatic), 7.53 (d, 1H, J = 4 Hz, aromatic). 2-(3,3-Dimethyl-6-hydroxyhexanoyl)-5-(5-phenyl-pentyl)thiophene (7). The nitrile (0.8 g, 2. 18 mmoles) was added to 20 ml of diethylene glycol containing KOH (1.25 g, 22 mmol) and 85% hydrazine hydrate (256 mg, 4 mmol) and the mixture was refluxed for 1 h. The mixture was distilled until the solution reached a temperature of 210deg.C and then heated under reflux for 3 h. After cooling to 27deg.C the reaction mixture was poured into 100 ml of H20, acidified to pH 3 with 12 M HCl, and extracted several times with Et2O. The combined Et2O extracts were washed thoroughly with H2O and dried over anhydrous MgSO4, and the Et2O was removed in vacuo to afford 2(3,3-dimethyl-6-hydroxyhexanoyl)-5-(5-phenyl-pentyl)thiophene (7; 742 mg, 92%) as a pale yellow oil; NMR (CDCl3) d 1.05 (s, 6H, (CH3)2), 1.5-1.9 (m, 10H, CH2), 2.3-3.0 (m, 8H, thienyl-CH2-, Ph-CH2-, CH2COOH), 6.6 (s, 2H, thienyl), and 7.27 (s, 5H, phenyl). 15-Phenyl-4,4-dimethylpentedecanoic acid(8). Raney nickel (5 g) and the thienyl acid 7 (0.6 g, 1.6 mmol) were vigorously stirred and refluxed in 100 ml of 10% Na2CO3 for 4 h. The hot solution was filtered through Celite, and the cooled filtrate carefully acidified to pH = 3 with 12 M HCI and extracted thoroughly with Et2O. The combined Et20 extracts were washed several times with H20, and dried over anhydrous MgS04, and the solvent was evaporated in vacuo to give an oil. The crude product was applied to a silicic acid (30 g) column slurried in C6H6. Fractions (20 ml) 1-lO were eluted with C6H6 followed by elution with 20 ml fractions of CH2Cl2. Fractions 11-20 were combined to give 15-phenyl-4,4'-dimethyl-pentadecanoic acid (8; 520 mg, 95%) as a colorless oil; NMR (CDCl3) d 1.05 (s, 6H, (CH3)2), 1.27 (s, 22H, CH2), 2.3 (m, 2H, CH2COOH), 2.6 (m, 2H, PhCH2), and 7.3 (s, 5H, aromatic); MS, M+ = m/z 346. 

15-(p-Iodophenyl)-4,4-dimethylpentadecanoic acid (9). 

The acid 8 was treated with thallium(III)trifluoroacetate followed by Kl as described previously (vide ante) to provide the para-iodo-substituted product 9, 4,4-DMIPP; NMR (CDC13) d0.82 (s, 6H, (CH3)2), 1.24 (m, ~22H, CH2), 2.30 (t 2H -CH2CO2- ) 2.59 (t, 2H PhCH2), and 7.24 (A2B2, 4H J=18 Hz) and MS, M+ = m/z 472. 

Synthesis of 15-(p-iodophenyl)-9, 9-dimethylpentadecanoic Acid (9,9-DMIPP; Scheme II) 2-(2'-Thienyl)-2-(5'-2-phenyl-1-oxo-ethyl-2'-thienyl)-propane (12). Phenacyl chloride (11; 4.06 g, 36 mmol) and 2,2-di(2'-thienyl)propane (10; 6.24 g, 30 mmol) in 125 ml of CH2C12 and anhydrous SnCl4 (10.4 g, 40 mmol) was 
reacted as described for 5. The resulting purple-colored solution was then treated with 6 M HCI until an amber-
colored solution was obtained. The CH2C12 layer was separated and thoroughly washed stepwise with 1O% HCl, H20, and 10% NaOH, and dried over anhydrous MgS04. Evaporation of the CH2Cl2 in vacuo afforded 12 as an orange-colored oil. The crude product was applied to a silica gel (200 g) column slurried in C6H6. Fractions (20 ml) 1-40 were eluted with C6H6. Fractions 19-31 were combined to give 5.73 g of 12 (63%) as light yellow-colored oil; NMR (CDCl3) d 1.8 (s, 6H, (CH3)2), 3.3 (s, 2H, PhCH2C=O), 6.4-7.6 (m, 5H, thienyl), 7.4 (s, 5N, aromatic). 
 2-(2'-Thienyl)-2-(5'-2-phenylethyl-2'-thienyl)propane (13). The ketone 12 (4.5 g, 15 mmol) was added to 40 ml of diethylene glycol containing KOH (6 g, 107 mmol) and 85% hydrazine hydrate (3 g, 48 mmol) and the mixture 
was reacted as described for 6. After cooling to 27deg.C the reaction mixture was poured into 300 ml of H2O and extracted several times with Et2O. The combined Et2O extracts were washed thoroughly with H2O and dried over anhydrous MgSO4, and the Et2O was removed in vacuo to give 13 as a yellow-colored oil. The crude product was applied to column containing silica gel (1OO g) slurried in petroleum ether. Fractions (20 ml) 1-40 were eluted with petroleum ether. Fractions 35-60 were combined to give 2-(2'-thienyl)-2-(5'-2-phenylethyl-2'-thienyl)propane (13; 2.51 g, 60%) as a colorless oil; NMR (CDCl3) d 1.8 (s, 6H, (CH3)2), 3.0 (s, 4H, CH2-CH2), 6.4-7.2 (m, 5H, thienyl). 2-(2'-l-0xo-4-carbomethoxybutanoyl-2'-thienyl)-2-(5'-2-phenylethyl-2'-thienyl)propane (15). 4-Carbo-methoxypropionyl chloride (14; 2.0 g,12 mmol) and 2-(2'-thienyl)-2-(5'-2-phenylethyl-2'-thienyl)propane (13; 2.2 g, 8 mmol) in 80 mL of CH2Cl2 and anhydrous SnCl4 (6 g, 24 mmol) was reacted as described for 5. The resulting purple-colored solution was then treated with 6 N HCI until an amber solution was obtained. The CH2Cl2 layer was separated and thoroughly washed with 1O% HCl, H2O, 1O% NaOH and dried over anhydrous MgSO4. Evaporation of the CH2Cl2 in vacuo afforded 12 as an orange-colored oil. The crude product was applied to a silica gel (150 g) column slurried in C6H6. Fractions (20 ml) 1-40 were eluted with C6H6. Fractions 25-38 were combined to give 1.08 g of 15 (27%) as light yellow oil; NMR (CDCl3) D 1.8 (s, 6H, (CH3)2), 2.7-3.4 (m, 8H, Ph-CH2CH2 , C=O-CH2-CH2 ), 3.67 (s, 3H, COOCH3), 6.4-7.6 (m, 4H, thienyl), 7.4 (s, 5H, aromatic). 
 2-(2'-4-Hydroxybutyl-2'-thienyl)-2-(5'-2-phenylethyl-2'-thienyl)propane (16). The ketoester 15 (1.00 g, 2.5 mmol) was added to 30 mL of diethylene glycol containing KOH (750 mg, 13.5 mmol) and 85% hydrazine hydrate (500 mg, 8 mmol) and the mixture was reacted as described for 6. After cooling to 27C the reaction mixture was poured into 400 ml of H2O acidified to pH 3 with 12 M HCl, and extracted several times with Et2O. The combined Et2O extracts were washed thoroughly with H2O and dried over anhydrous MgSO4, and the Et2O was removed in vacuo to give 16 as a yellow-colored oil. The crude product was applied to a silica gel (30 g) column slurried in CH2Cl2. Fractions (20 ml) 1-20 were eluted with CH2Cl2. Fractions 4-1O were combined to give acid 16 (435 mg, 45%) as a colorless oil; NMR (CDCl3) D 1.8 (s, 6H, (CH3)2), 1.5-1.9 (m, 2H, CH2-CH2COON), 2.5-2.9 (m, 8H, Ph CH2 CH2 , thienyl-CH2-, -CH2COOH), 6.6 (d, 2H, thienyl), 6.9 (d, 2H, thienyl), 7.4 (s, 5H, aromatic). 15-Phenyl-9,9-dimethylpentadecanoic acid (17). Raney nickel (35 g) and the thienyl acid (16; 390 mg, 1 mmol) were reacted in 300 mL of 1O% Na2CO3 containing 30 ml ethanol as described for 8. The filtered Raney nickel was cooled and carefully acidified to pH 3 with 12 M HCl and extracted thoroughly with Et2O. The filtered Raney 
nickel was dissolved in 12M HCl and the resulting green solution was extracted with Et2O as described above. The combined Et2O extracts were washed several times with H2O, and dried over anhydrous MgSO4 and the solvent was evaporated in vacuo to give an oil. The crude product was applied to a silicic acid (30 g) column slurried in C6H6. Fractions (20 ml) 1-10 were eluted with C6H6 followed by elution with 20 ml fractions of CH2Cl2. Fractions 9-16 were combined to give 15-phenyl-9,9-dimethyl-pentadecanoic acid (17; 200 mg, 60%) as a colorless oil; NMR (CDCl3) D 0.8 (s, 6H, (CH3)2), 1.0-1.8 (brs, 22H, CH2), 2.34 (t, 2H, J = 4 Hz, CH2COOH), 2.65 (t, 2H, J = 7 Hz, Ph-CH2), 7.3 (s, 5H, aromatic); MS, M+ = m/z 346 
 15- (p- lodophenyl-9, 9-dimethylpentadecanoic acid (18). The acid substrate 17 was converted to the paraiodo product 18 in the usual manner (vide ante); NMR (CDCl3) D 0.799 (s, 6H, (CH3)2), 1.23 (m, 22H, -CH2) and 7.24 (A2B2, 4H, J = 18 Hz); MS, M+ = m/z 472. 

Synthesis of 15-(p-iodophenyl)-6, 6-dimethyl-pentade-canoic Acid (6,6-DMIPP: Scheme III) 2-(2' thienyl)-2-(5' carboxy-2'-thienyl)propane (19). The 2,2-di(2'-thienyl)propane (10; 10.4 g, 0.05 mol) was stirred in 100 ml of dry diethyl ether under nitrogen at ambient temperature and 40 ml of a 1.55 M n-butyllithium solution was added with a syringe through a rubber septum. The resulting mixture was stirred at ambient temperature for 15 min and poured onto a mixture of diethyl ether and crushed dry ice. The resulting solution was poured into 100 ml of H2O and extracted several times with diethyl ether. The diethyl ether extracts were washed with 50 ml of H2O and the combined H2O phases were acidified to pH 3 with 12 M HCl and extracted several times with diethyl ether. The combined diethyl ether extracts were washed several times with water, dried over anhydrous Na2SO4 and the solvent was evaporated in vacuo to yield 14.8 g of crude 2-(2'-thienyl)-2-(5'-carboxy-2'-thienyl)propane (19); NMR (CDCl3) 8 1.9 (s, 6H, (CH3)2), 6.8-7.6 (m, 5H, thienyl).  
 2- (2'-Thienyl)-2-(5'-carbomethoxy-2'-thienyl)propane (20). The acid 19 (14.8 g, 0.059 mol) was added to an ether solution (1OO ml) containing excess CH2N2, prepared from N-methyl-N'-nitro-N-nitrosoguanidine (MNNG; 7.5 g). The mixture was stirred at 0deg.C under red lights for 12 h and the Et2O solution dried over anhydrous Na2SO4, and the solvent was removed in vacuo to yield an oil. The crude product was applied to a silica gel ( 100 g, Davidson) column slurried in petroleum ether (30-60deg.C boiling range). Fractions (20 ml) were eluted with 1 : l petroleum ether (30-60deg.C boiling range): benzene. Fractions 20-50 were combined to give 2-(2'-thienyl)-2-(5'-carbomethoxy-2'-thienyl)propane (20; 4.73 g, 37%) as a pale yellow oil; NMR (CDCl3) d 1.9 (s, 6H, (CH3)2), 3.65 (s, 3H, -COOCH3), 6.8-7.6 (m, 5H, thienyl). 
 2- (5'-5-Phenyl-l-oxo-pentyl-2'-thienyl)-2-(5'-carbo-methoxy-2'-thienyl)propane (21). Anhydrous SnC14 (5.2 g, 20 mmol) was added dropwise to a solution of 5-phenylpentanoyl chloride (4; 2.0 g, 10 mmol) and 2-(2'-thienyl)-2-(5'-carbomethoxy-2'-thienyl)probane (20; 2.66 g, 10 mmol) in CH2Cl2 (1OO ml) stirred at Odeg.C. The resulting purple solution was stirred at Odeg.C for 30 min and then at room temperature for 2 h and treated with 150 ml of 6 M HCl at 0-5deg.C until a yellow-colored solution was obtained. The CH2Cl2 layer was washed (4 x 100 ml) with 1 M HCl, with H2O (3 x 100 mL), dried over anhydrous Na2SO4 and the CH2Cl2 was removed in vacuo to afford 4.31 g of 21 as a yellow oil; NMR (CDCl3), 1.5-1.9 (m, 4H, Ph-CH2-CH2-CH2-CH2-C=O), 1.9 (s, 6H, (CH3)2), 
2.4-2.9 (m, 4H, Ph-CH2-, CH2-C=O), 3.65 (s, 3H, COOCH3), 6.85 (d, 2H, J = 4 Hz, thienyl), 7.2 (s, 5H, aromatic), 7.45 (d, 1H, J = 4 Hz, thienyl), 7.6 (d, 1H). 2 - (5'-5- phenylpentyl- 2'-thienyl) - 2 - (5'-carboxy-2'-thienyl)propane (22). The acid 21 (3.2 g, 7.5 mmoles) was added to 60 ml of diethylene glycol containing KOH (1.5 g, 27 mmol) and 85% hydrazine hydrate (1 g, 30 mmol) and the mixture refluxed for 1 h. The mixture was distilled until the solution reached a temperature of 210deg.C and then heated under reflux for 3 h. After cooling to 27deg.C the reaction mixture was poured into 300 ml of H20, acidified to pH 3 with 12 M HCl, and extracted several times with Et2O. The combined Et2O extracts were washed thoroughly with H20 and dried over anhydrous Na2S04, and the Et2O removed in vacuo to afford 2-(5'-5-phenyl-pentyl-2'-thienyl)-2-(5'-carboxy-2-thienyl)propane (22; 2.84 g, 94%) as a pale yellow oil; NMR (CDC13) d 1.4-1.7 (m, 6H, Ph-CH2-CH2-CH2-CH2), 1.85 (s, 6H, (CH3)2), 2.5-2.9 (m, 4H, Ph-CH2-,thienyl-CH2-), 6.55 (d, 1H, J = 4 Hz, thienyl), 6.7 (d, 1H, J = 4 Hz, thienyl), 6.85 (d, 1H, J = 4 Hz, thienyl), 7.2 (s, 5H, aromatic), 7.8 (d, 1H, J = 4 Hz, thienyl). 
 15-Phenyl-6, 6-dimethylpentadecanoic acid (23). Raney nickel (80 g) and the thienyl acid 22 (2.0 g, 5 mmol) were vigorously stirred and refluxed in 300 ml of 1O% Na2CO3 for 10 h. The hot solution was filtered through Celite, and the cooled filtrate carefully acidified to pH 3 with 12 M HCl and extracted thoroughly with Et2O. The combined Et2O extracts were washed several times with H2O, and dried over anhydrous MgSO4, and the solvent was evaporated in vacuo to give 600 mg of an oil which crystallized on standing. The filtered Raney nickel was dissolved in 12 M HCl and the resulting green-colored solution extracted with Et2O as described above to give an additional 425 mg of 15-phenyl-6,6-dimethylpentade-canoic acid (23), 59% yield as a clear yellow-colored oil; NMR (CDC13) d 0.8 (s, 6H, (CH3)2), 1.0-1.8 (s, 22H, CH2), 2.3-2.8 (m, 4H, Ph-CH2, CH2COOH), 7.2 (s, 5H, aromatic); MS, M+ = m/z 472. 15-(p-Iodophenyl)-6, 6-dimethylpentadecanoic acid (24). The acid 23 was converted to 24 in the usual manner (vide infra); NMR (CDC13) 6 0.81 1 (s, 6H, (CH3)2), 1 24 (m 22H CH2), 2.35 (t, 2H, -CH2C02-), 2.51 (t, 2H, Ph-CH2), and 7.24 (A2B2, 4H, J = 1 8 Hz); MS, M+ = m/z 472. 

 Preparation of 2-(3, 3)-Dimethyl-1 -hydroxypentanoyl)-Thiophene (1) Substrate and Preparation of 3, 3-Dimethyl Analoques 28a-28f 

This substrate was used for the synthesis of the 3,3-dimethyl-substituted analogues 28a-28f and was prepared as described earlier by the acylation of thiophene with the monomethyl ester of 3,3-dimethylglutaric acid followed by Wolff-Kishner reduction (Knapp et al., 1986a). 

Synthesls of 15-(p-iodophenyl)-3,3-dimethylpentade-canoic Acid (3,3-DMIPP, 28e) The 3,3-dimethyl analogue (Fig. l) was prepared as described earlier*, m.p. 37-39'C; MS, M+ = m/z 472. 

Synthesis of 3,3-Dimethyl Analogues of 15-phenylpenta-decanoic Acid (28a-28f Scheme 1V) These analogues (Scheme IV) were synthesized by Freidal-Craft acylation of the thiophene template (1) with the appropriate terminal phenyl-substituted acyl chloride followed by Wolf-Kishner reduction and para-iodination, as described in Table 1 . The acyl chlorides, prepared from the commercially available w-phenylalkanoic acids, used for the various analogues were as follows: 28a, phenylacetic acid; 28b, 3-phenylpropionic acid; 28c, 4-phenylbutyric acid; 28d, 5-phenylpentanoic acid; 28e, 6-phenylhexanoic acid; 28f, 10-phenyldecanoic acid. 

 Preparation and Formulation of Iodine-125-labeled Fatty Acids Iodine-125 was introduced into the para-position of the ternrinal phenyl-substituted analogues in the usual manner by potassium iodide treatment of the thallated intermediate prepared by reaction of the substrate with 2 equivalents of thallium(III)trifluoroacetate in trifluoroacetic acid in the dark overnight. All radioiodinated fatty acid analogues had similar specific activity values of 2.4-4 Ci/mmole. The fatty acid was dissolved in a minimal amount of warm ethanol (-lOO ul) which was added slowly with stirring to a 6% solution of bovine serum albumin at 45-50deg.C. The solution was filtered through a Millipore filter before injection. 

Tissue Distribution Studies 

The distribution of radioactivity was determined in tissues of 6-8 week-old female Fischer 344 rats ( 120-130 gm) after i.v. administration of the radioiodinated fatty acids. Food was removed from the rats 18 h prior to initiation of the experiment, but the animals were allowed water ad libitum prior to and during the course of the experiment. The radioiodinated fatty acids were dissolved in absolute ethanol (-lOO ul) which was added dropwise to a stirred solution of 6% bovine serum albumin at -40deg.C. The final ethanol concentration was less than 10%. The solution was filtered through a 0.22-um Millipore filter and injected through a lateral tail vein into the diethyl ether-anesthetized animals. After the times indicated, the animals were killed and blood samples were obtained by cardiac puncture. The organs were then removed, rinsed with saline, and blotted dry to remove residual blood. The organs were weighed and counted in a Nal autogamma counter. For the dual-labeling experiments, the 123-I ( 159 keV) photopeak was counted and the samples then stored in the cold until the 123-I contribution to the 1251 x-ray photopeak region was < 4-5%. The samples were then counted again to determine the distribution of 125-I. Samples of the injected radioactive solutions were also assayed at both counting periods as standards for decay corrections and to calculate the percent injected dose per gm of tissue values. The thyroid glands were not weighed directly. The weight of the thyroid glands was calculated in the usual manner by multiplying the animal weight by 7.5 mg/lOO gm. 

RESULTS 
Chemical Syntheses of New Analogues 

Four dimethyl-branched analogues of Cardiodine were synthesized (Fig. 2). The 4,4-DMIPP analogue was synthesized in a 7-step reaction sequence (Scheme I) from a key synthetic intermediate, 2-(3,3-dimethyl-5-hydroxy-pentanoyl)thiophene (1), which was also utilized earlier for introduction of dimethyl-branching into the alkanoic chain for preparation of 3,3-DMIPP34 and 3,3-dimethyl-19-E-iodononadecanoic acid (3,3-DMIVN).35 In this synthetic approach, 2-(3,3-dimethyl-5-hydroxypentanoyl)-thiophene (1) was reduced with LiAlH4 to afford 2-(3,3-dimethyl-5-hydroxypentyl)thiophene (2). The thienyl alcohol (2) was converted to the alkyl chloride (3) by treatment with N,N'-dimethylaniline and thionyl chloride. The phenylalkyl moiety was then introduced into the 5'-thienyl position by coupling of the acid chloride of 5-phenylpentanoic acid (4) with chloride (3) catalyzed by treatment with tin(IV) chloride in methylene chloride. 
The chloroketone (5) was then converted to the nitrile (6) by treatment with NaCN in DMSO. The keto function of compound (6) was reduced by the Wolff-Kishner (Huang-Minlon) method which concomitantly hydrolyzed the nitrile function to give 2-(4,4-dimethyl-6-hydroxy-hexanoyl)-5-(5-phenylpentyl)thiophene (7). Raney Nickel desulfurization of the thienyl acid (7) to compound (8) followed by subsequent treatment with thallium(III)-trifluoroacetate and potassium iodide gave 4,4-DMIPP (9). 
 For synthesis of the 6,6- and 9,9-dimethyl analogues, dimethyl-branching was introduced utilizing bis(thienyl)-propane (10, Scheme II) with subsequent chain elongation via fabrication of the 5,5'-disubstituted thiophene system by successive Friedal-Crafts and Wolff-Kishner sequences. By selection of the substituents introduced into the 2- and 2'-positions of the thiophene ring of bis-(thienyl)propane (10), a variety of dimethyl-branched fatty acids with branching at positions 6 through 10 can be prepared. In the synthetic approach for preparation of these analogues, bis(thienyl)propane (10) was prepared by treatment of thiophene and acetone with 75% sulfuric acid. The synthesis of the analogue 9,9-DMIPP (18) was accomplished in 9-steps as shown in Scheme II. Utilizing this approach, commercially available phenylacetyl chloride (11) was coupled with bis(thienyl)propane (10) by treatment with tin(IV) chloride in dichloromethane which gave 2-(2'-thienyl)-2-(5'-2-phenyl-1-oxo-ethyl-2'-thienyl)propane (12). Wolff-Kishner reduction of 1 2 gave 2-(2'-thienyl)-2-(5'-2-phenylethyl-2-thienyl)propane (13). The carboxylic acid moiety was introduced into the 2'-thienyl position by acylation of the half ester acid chloride of succinic anhydride. The 3-carbomethoxypropionyl chloride ( 14) was prepared from commercially available succinic anhydride by treatment with methanol followed by thionyl chloride. The thienyl derivative (13) was subjected to Friedel-Crafts condensation with the acyl chloride (14) to afford 2-(2'-1-oxo-4-carbomethoxy-butanoyl-2'-thienyl)-2-(5'-2-phenylethyl-2'-thienyl)-pro-pane (15). Wolff-Kishner reduction of the keto ester (15) gave 2-(2'-4-hydroxybutyl-2'-thienyl)-2-(5'-2-phenyl-ethyl-2'-thienyl)propane (16). The pivotal step in the synthesis involved the Raney nickel desulfurization of 16 to provide 17, followed by subsequent treatment with the thallium(III)trifluoroacetate and potassium iodide to give 9,9-DMIPP (18). 
 The preparation of 15-(p-iodophenyl)-6,6-dimethyl-pentadecanoic acid (6,6-DMIPP) was achieved in a manner analogous to the 9,9-analogue as shown in Scheme 111. The 2,2-di(2'-thienyl)propane (10) was converted to the 2-lithio derivative using n-butyllithium and then reacted with dry ice to give 2-(2'-thienyl)-2-(5'-carboxy-2'-thienyl)propane (19). Compound 19 was then treated with diazomethane to give the corresponding methyl ester (20). The resulting ester and 5-phenylpentanoyl chloride (4) were subjected to Friedel-Crafts acylation to afford 2-(5'-5-phenyl-1-0xo-pentyl-2'-thienyl)-2-(5'-car-bo-methoxy-2'-thienyl)propane (21). Wolff-Kishner reduction of ketone (21) gave 2-(5'-5-phenylpentyl-2'-thienyl)-2-(5'-carboxy-2'-thienyl)propane (22). Raney nickel desulfurization of the thienyl acid (22) and subse-quent iodination using thallium(III)trifiuoroacetate and potassium iodide afforded 6,6-DMIPP (24). 
 The 3,3-dimethyl analogues 28a-28f (Fig. 3) were synthesized by acylation of 2-(3,3-dimethyl-5-hydroxy-pentyl)thiophene (1 , Scheme IV). Five 3,3-dimethyl ana-10gues (Fig. 1 ) with C-1l, C-12, C-13, C-14, C-15 (the original 3,3-DMIPP analogue reported earlier; Knapp et al., ref. #34) and C-19 carbon chain lengths were synthe-sized from the common 2-(3,3-dimethyl-1-hydroxy-pentanoyl)-thiophene intermediate (1 , Scheme IV). By choice of the appropriate w-phenylalkanoyl chloride, the various analogues were readily prepared by acylation of (1) followed by Wolff-Kishner deoxygenation of the disubstituted thiophene product (25) and sulfur extrusion of (26) with Raney Nickel. Iodine was then introduced into the para-position of (27) by the usual thallationpotassium iodide route to provide the final product (28). The iodine-125-labeled analogues were prepared for tis-sue distribution studies in fasted rats in the same manner using iodine-125. 
Biological Studies-Effects of Position of Dimethyl 
Branching on 15-Carbon Chain Length Analogues The 3,3-, 4,4-, 6,6-, and 9,9-DMIPP analogues were radioiodinated with [iodine-125] sodium iodide and the radioiodinated agents evaluated in fasted female Fisher rats. The results of the biological evaluation are summarized in Table 2. The level of accumulation of radioactivity in the myocardium after injection of 6,6-and 9,9-DMIPP was only moderate. The blood levels of activity were relatively high resulting in modest heart: blood ratios (< 3 : l). These analogues thus appear to show lower heart uptake, more rapid myocardial washout, and higher blood levels than for the 3,3-DMIPP analogue. Since these analogues all have the same C15 chain length, the results indicate that the position of dimethyl-branching is an important structural feature. In contrast, both the 3,3-DMIPP and 4,4-DMIPP analogues exhibit excellent myocardial retention (> 95%) 30 min following intravenous injection, which decreases to < 65% after 60 min (Fig. 4). 

Enects of Chain Length of 3,3-Dimethyl Analogues 

In a similar manner to the comparative evaluation of the 3,3-, 4,4-, 6,6-, and 9,9-dimethyl analogues of 15-(p-iodophenyl)-pentadecanoic acid, the six 3,3-dimethyl analogues with chain lengths varying from I 1-19 carbon atoms (28a-28D were also evaluated in fasted rats (Table 4). This series of experiments demonstrated that the total chain length is also an important feature, with the 15-carbon chain length 3,3-dimethyl analogue (28e) showing the highest myocardial extraction and resulting in the highest heart: blood ratios (Fig. 5). Consistent with the higher in vivo uptake of the 4,4-DMIPP analogue (9) observed in rats in comparison with 3,3-DMIPP (28e), other studies with the traditional non-working Langendorff-perfused isolated rat heart preparation have demonstrated good uptake in vitro and nearly irreversible retention with only very low subsequent loss of activity in the outflow (unpublished data). 

DISCUSSION 

The presence of dimethyl-branching in the 3-position of radioiodinated DMIPP34 has been shown to result in significantly increased retention of radioactivity after intravenous administration to rats (3,3-DMIPP; Table 2). The results of the individual studies summarized in Table 2 appear to indicate that the 4,4-DMIPP analogue has greater heart uptake and higher heart: blood ratios than the model 3,3-DMIPP analogue. In addition, radioiodinated 3,3-DMIPP also shows much longer retention in the canine heart in comparison with the unbranched IPPA and 3-(R,S)-monomethyl-branched BMIPP analogues. (1-,34) In addition to the studies described above, the subcellular and lipid pool distribution of radioiodinated 3,3-DMIPP have also been evaluated in detail in vivo following administration to Fischer rats 8 and in the endogenous lipids of Langendorff perfused rat hearts. The high myocardial extraction and longer retention of radioiodinated DMIPP in comparison with Cardiodine determined in these studies appears to be related to the slow conversion to triacylglyceride storage products. More recently, the high myocardial uptake and prolonged retention was further confirmed in the normoxic hearts of a canine model and the incorporation into endogenous lipids has also been reported,37 clearly demonstrating that geminal 3,3-dimethyl substitution does not interfere with targeting of selected long chain fatty acids to the myocardium. Both 3,3-DMIPP and 4,4-DMIPP are thus good candidates for further evaluation. 
 Although the exact mechanism of retention has not yet been elucidated, analogues which show retention such as 3,3- and 4,4-DMIPP are presumed to be incorporated into myocardial triglyceride storage products, as has been demonstrated in earlier studies with 3-R,S-BMIPP and 3,3-DMIPP.1-5,8,16 In addition, the mechanisms involved in myocardial uptake of fatty acids and the involvement of fatty acid binding proteins on the transfer from the interstitial space and through the myocyte membrane are not well understood. The various analogues also show vary-ing degrees of deiodination expressed as the per cent injected dose per gram of tissue values (Tables 3 and 4). Although the reasons for these differences are not well understood, when expressed as per cent injected dose values, the deiodination in all cases is very low. The results of studies reported by other investigators with I -[11C]-3,3-dimethylheptadecanoic acid (DMHDA) have suggested that 3,3-dimethyl branching decreases myocardial uptake and retention in comparison with the analogues mono-branched analogues.1-8 In fasted rats (n = 4) the DMHDA analogue showed relatively low myocardial uptake and relatively rapid washout (mean % injected dose/gm heart, heart : blood = H/B ratio): 5 min, 0.63, H/B 1.19 : 1; 15 min, 0.46, H/B 0.94; 30 min, 0.42, H/B 1.08.38 In another study, [1-125]-14-iodo-3,3-dimethyltetradecanoic acid (IDTDA), also showed de-creased myocardial uptake and retention, although no systematic results were reported to evaluate the effects of chain length.39 In comparison with our results with 3,3-DMIPP and several other analogues described in the present work, the behavior of DMHDA and IDTDA is unexpected, but indicates that other structural features in addition to dimethyl-branching must affect myocardial extraction and release kinetics. These results demonstrate that both DMHDA (PET) and IDTDA (SPECT) would not be expected to be good candidates for tomography because of the low heart uptake, low target/non-target ratios and relatively rapid myocardial washout. 
 Our systematic evaluation with the nine new dimethyl-branched terminal p-iodophenyl-substituted analogues in comparison with 3,3-DMIPP has clearly shown that geminal 3,3-dimethyl substitution in itself does not necessarily decrease either myocardial specificity or retention and demonstrate that both chain length and the position of dimethyl branching are important structural features. There are a variety of factors which may affect the unusual behavior of DMHDA which may in fact include an oxidative decarboxylation process, since the possibility of expired 11CO2 was evidently not measured by these investigators. In addition, fatty acid transport of DMHDA may be uniquely impaired by a combination of structural features, i.e. total chain length in conjunction with 3,3-dimethyl substitution. Geminal 3,3-dimethyl substitution alone does not reduce the myocardial specificity of 3,3-dimethyl substituted long chain fatty acids. These agents, if appropriately designed, can result in very high heart uptake and quite prolonged retention, as evidenced by both 3,3-DMIPP and 4,4-DMIPP in the current studies. The current study focused on detailed structure-activity studies with ten dimethyl-branched fatty acid ana-logues in animals. Both 3,3-DMIPP (28c) and 4,4-DMIPP (9) analogues show the best myocardial uptake and retention. These combined results conclusively demonstrate that dimethyl-substitution can be an effective structural modification which can significantly prolong myocardial residence and that iodine- 123-labeled 3,3-DMIPP (28e) and 4,4-DMIPP (9) are candidates for more detailed future SPECT studies in humans. ACKNOWLEDGMENTS 

Research at the Oak Ridge National Laboratory sponsored by the Office of Health and Environmental Research, U.S. Department of Energy under contract DE-AC05-96OR22464 with Lockheed Martin Energy Research Corporation. The authors also thank the National Institutes of Health (#HL 35500, F.F.K.), the North Atlantic Treaty Organization Collaborative Grants Programme (CRG 900966, F.F.K. and J.K.), and the Deutsche Forschungs Gemeinschaft (J.K.), and F.F. Knapp, Jr. also ac-knowledges support from the Alexander von Humboldt Foundation of Germany for support for the 1991-1992 period at the University of Bonn, Germany. The authors also thank D.E. Rice and E.C. Cunningham for technical assistance and Ms. L.S. Ailey for secretarial assistance. 

REFERENCES 

l . Knapp Jr FF, Kropp J. Goodman MM, Franken P, Reske SN, Ambrose, KR, et al. The development of iodine-123-methyl-branched fatty acids and their applications in nuclear cardiology. Ann Nucl Med 7: SII 1-14, 1993. 
2. Knapp FF Jr, Goodman MM. The design and biological properties of iodine-123-labeled 3-methyl-branched fatty acids. In Proceedings of the Workshop on Radiolabeled Free Fatty acids, Academic Hospital Free University, Amsterdam, The Netherlands, July 6, 1984; Eur Heart J 6: 71-84, 1985. 
3. Knapp FF Jr, Goodman MM, Ambrose KR, Som P, Brill AB, Yamamoto K, et al. The development of radioiodinated 3-methyl-branched fatty acids for evaluation of myocardial disease by single photon techniques. In Noninvasive Measurement of Cardiac Metabolism, EE van der Wall (ed.), Amsterdam, Martinus Nijhoff Publishers, pp. 159-202, 1987. 
4. Knapp JR FF, Ambrose KR, Goodman MM. New radioiodinated methyl-branched fatty acids for cardiac imaging. Eur J Nucl Med 12: S539-544, 1986. 
5. Knapp Jr FF, Kropp J. Iodine-123-labeled fatty acids for myocardial SPECT: Current status and future perspectives. Eur J Nucl Med, Invited Review, 22: 361-381 , 1995. 
6. Dudczak R, Schmoliner R, Angelberger P, Knapp FF Jr, Goodman MM. Structurally modified fatty acids: clinical potential as tracers of metabolism. Eur J Nucl Med 12: S45-S48, 1986. 
7. Goodman MM, Callahan AP, Cunningham EB, Kirsch G, Knapp FF Jr. Synthesis and evaluation of radioiodinated terminal p-iodothienyl-substituted a- and b-methyl-branched fatty acids: a new class of myocardial imaging agents. J Med Chem 27: 390-396 1984. 
8 Ambrose KR, Owen BA. Goodman MM, Knapp FF Jr. Evaluation of the metabolism in rat hearts of two new radioiodinated 3-methyl-branched fatty acid myocardial imaging agents. Eur J Nucl Med 12: 486-491 , 1987. 
9. Fischman AJ, Ahmad M, Wilkinson R, Keech F, Saito T, Strauss HW. The role of combined fatty acid and thallium imaging in the evaluation of myocardial ischemia and infarction. J Nucl Med 29 (5): 843 (abst #423), 1988. 
10. Yamamoto K, Som P, Brill AB, Yonekura Y, Srivastava SC, Meinken GE, et al. Dual tracer autoradiographic study of B-methyl(14C)heptadecanoic acid and 15-p-(131I)-iodophenyl-B-methylpentadecanoic acid in normotensive and hypertensive rats. J Nucl Med 27: 1178-1183, 1966. 
11 . Yonekura Y, Brill AB, Som P, et al. Quantitative autoradiographic measurement of regional myocardial substrate utilization in hypertensive rats. Science 227: 1494-1496, 1985. 
12. Nishimura T, Sago M, Kihara K, Oka H, Shimonagata T, Katabuchi T, et al. Fatty acid myocardial imaging using 121I-B-methyl-iodophenyl pentadecanoic acid (BMIPP): comparison of myocardial perfusion and fatty acid utilization in canine myocardial infarction (occlusion and reperfusion model). Eur J Nucl Med 15: 341-345, 1989. 
13. Fujibayashi Y, Yonekura Y, Takemura Y, Wada K, Matsumoto K, Tamaki N, et al. Myocardial accumulation of iodinated beta-methyl-branched fatty acid analogue, iodine-125-15-(p-iodophenyl)-3-(R,S)-methylpentadecanoic acid (BMIPP), in relation to ATP concentration. J Nucl Med 31 : 1818-1822, 1990. 
14. Fujibayashi Y, Yonekura Y, Tamaki N, Yamamoto K, Som P, Knapp Jr FF, et al. Myocardial accumulation of BMIPP in relation to ATP concentration. Ann Nucl Med 7: SII 15-18, 1993. 
15. Knapp FF Jr, Reske SN, Ambrose KR, Kohlen S, Kolkmeier J, Goodman MM, et al. Formation of polar products from radioiodinated 15-(p-iodophenyl)-3-(R,S)-methylpentade-canoic acid (BMIPP) by isolated Langendorff rat hearts. 
NucCompact-European/American Communications in Nuclear Medicine 21 : 133-139, 1990. 
16. Knapp Jr FF, Goodman MM, Reske SN, et al. Radioiodinated methyl-branched fatty acids-evaluation of catabolites formed in vivo. NucCompact/EurAmer Commun Nucl Med 21 : 229-231 , 1990. 
17. Kropp J, Knapp FF Jr, Ambrose KR, Rogers CJ, Callahan AP, Biersack H-J. Identifizierung eins unerwarteten metaboliten von 15-(p-iodophenyl)-3-R,S-methylpentadecansaure (BMIPP) am isolierten rattenherzen. J German Soc Nucl Med p 54 (abst V116), 1989. 
18. Kropp J, Knapp FF Jr, Assman T, Urmetz S, Nissen HP, Ambrose KR, et al. Metabolites of IPPA, BMIPP and DMIPP fatty acids in rat hearts: a quantitative study. Eur J Nucl Med 16: 408 (abstract), 1990. 
19. Robbins MS, Adams MD, Anderson HA, Dean RT, White DH. Myocardial uptake of radioiodine-labeled 15-(4-iodophenyl)-9-methylpentadecanoic acid in laboratory animals. J Nucl Med 26: P124 (abst #530), 1985. 
20. Miller DD, Gill JB, Barlai-Kovach M, Nedelman M, Livni E, Elmaleh DR, et al. Does modified fatty acid analog 'redistribution' occur in ischemic myocardium? J Nucl Med 27: 966 (abst #367), 1986. 
21. Chouraqui P, Maddahi J, Henkin R, Karech SM, Galie E, Berman DS. Comparison of myocardial imaging with iodine- 123-iodophenyl-9-methyl pentadecanoic acid and thallium-201 -chloride for assessment of patients with exerciseinduced myocardial ischemia. J Nucl Med 32: 447-452, 1991. 
22. Nishimura T, Uehara T, Shimonagata T, Nagata S, Haze K. Clinical experience of 123-I-BMIPP myocardial imaging for myocardial infarction and hypertrophic cardiomyopathy. Ann Nucl Med 7: SII 35-4O, 1993. 
23. Takeishi Y, Chiba J, Abe S, et al. Heterogeneous myocardial distribution of iodine-123 15-(p-iodophenyl)-3-R,S-methylpentadecanoic avid (BMIPP) in patients with hypertrophic cardiomyopathy. Eur J Nucl Med 19: 775-782, 1992. 
24. Morita K, Yanagimoto S, Otsuka N, Ono S, Nagai K, Tomomitsu T, et al. I-123-BMIPP scintigraphy in seven cases with cardiomyopathy. Ann Nucl Med 7: SII 101-107, 1993 . 
25. Kurata C, Tawarahara K, Okayama K, Wakabayashi Y, Kobayashi A, Yamazaki N, et al. Myocardial imaging with radioiodinated beta-methyl-branched fatty acid in cardiomyopathy. Ann Nucl Med 7: SII 27-34,1993. 
26. Tamaki N, Kawamoto M, Yonekura Y, et al. Regional metabolic abnormality in relation to perfusion and wall motion in patients with myocardial infarction: Assessment with emission tomography using iodinated branched fatty acid analogue. J Nucl Med 33: 659-667, 1992. 
27. Tomiguchi S, Oyama Y, Nabeshima M, Nakashima R, Ono M, Kojima A, et al. Quantitative evaluation of BMIPP in patients with ischemic heart disease. Ann Nucl Med 7: SII 107-1 12, 1993. 
28. Kropp J, Juergans M. Glaenzer K, Luederitz B, Biersack H-J, Knapp Jr FF. Evaluation of ischemia and myocardial viability in patients with coronary artery disease (CAD) with iodine-123 labeled 15-(p-iodophenyl)-3-R,S-methyl-pentadecanoic acid (BMIPP). Ann Nucl Med 7: SII 93-1OO, 1993 . 
29. Franken PR, DeGeeter F, Dedale P, et al. Iodine-123 B-methyliodophenylpentadecanoic acid and Tc-99m-MIBI to identify ischemic but viable myocardium, In Proceedings of the First International Congress on Nuclear Cardiology, Cannes, France, April 25-28, 1993. 
30. DeGeeter FF, Franken P, Knapp, FF Jr, Boosuyt A. Relationship between blood flow and fatty acid metabolism in subacute myocardial infarction: a study by means of Tc-99m-Sestamibi and iodine-123-beta-methyl iodophenyl-pentadecanoic acid. Eur J Nucl Med 21 : 283-291 , 1994. 
31. Franken PR, De Geeter F, Dendale P, Demoor D, Block P, Bossuyt A. Abnormal free fatty acid uptake in subacute myocardial infraction after coronary thrombolysis: Correlation with wall motion and inotropic reserve. J Nucl Med 35: 1758-1765, 1994. 
32. Tamaki N, Kawamoto M, Yonekura Y, Fujibayashi Y, Magata Y, Torizuka T, et al. Assessment of fatty acid metabolism using I-123 branched fatty acid: Comparison with positron emission tomography. Ann Nucl Med 7: SII 41-48, 1993. 
33. Kawamoto M, Tamaki N, Yonekura Y, Tadamura E, Fujibayashi Y, Magata Y, et al. Combined study with I-123 fatty acid and thallium-201 to assess ischemic myocardium: Comparison with thallium redistribution and glucose metabolism. Ann Nucl Med 8: 847-854, 1994.
34. Knapp FF Jr, Goodman MM. Callahan AP, Kirsch G. Radioiodinated 15-(p-iodophenyl)-3,3-dimethylpentade-canoic acid: a useful new agent to evaluate myocardial fatty acid uptake. J Nucl Med 27: 521-531, 1986. 
35. Goodman MM, Neff KH, Ambrose KR, Knapp FF Jr. (E)-19-[125l]iodo-3,3-dimethyl-18-nonadecenoic acid: a new imaging agent to evaluate regional myocardial fatty acid uptake. J Nucl Med 28 (4): 724-725 (abst #711), 1987. 
36. Goodman MM. Goudonnet A, Knapp FF Jr. The position of geminal dimethyl-substitution affects myocardial uptake and clearance kinetics of DMIPP analogues in fasted rats. J Nucl Med 28 (4): 571 (abst #61), 1987. 
37. Visser FC. Duwel CMB, van Eenige MJ, Roos JP, Knapp FF Jr, van der Vusse GJ. Biochemistry of radioiodinated free fatty acids in the canine myocardium. J Cell Biochem 88: 185-190, 1989. 
38. Jones GS. Livni E, Strauss HW, Hanson RN, Elmaleh DR. Synthesis and biological evaluation of l-[C-1l]-3,3-dimethylheptadecanoic acid [(C-11)DMHDA]. J Nucl Med 29: 68-72, 1988. 
39. Otto CA, Brown LE, Lee H. Subcellular distribution of [125I]iodoaryl B-methyl fatty acids. Int J Nucl Med Biol 12 (3): 223-226, 1984.