electrophilic metal center has also been realized.  An example of such a reaction is shown in equation 3, whereby an anisole moiety bound in a cationic metal complex undergoes nucleophilic attack by the tert-butoxide ion to give tert-butyl methyl ether, and an uncharged complex having a coordinated phenolate anion [7].  This reaction occurs at ambient temperature in THF solvent.

            Another approach to effecting carbon-oxygen bond cleavage is to initiate nucleophilic attack at the electrophilic carbon center.  Subsequent hydrolysis can be used to regenerate the nucleophile (scheme A). Such reactions can be facilitated if the cation introduced with the nucleophile binds to the

ether oxygen, thereby increasing the electrophilicity of the ether carbons.  Precedence for such a pathway is found in the catalyzed hydrolysis of ethylene oxide to ethylene glycol.  Anionic catalysts such molybdate and sulfate are used, with the PPN salts being used for phase transfer catalysis [8].  The proposed pathway (scheme B) involves initial nucleophilic attack by the molybdate anion at the carbon

center. An alternative final step could involve closure to give a five-membered metallocycle that can undergo reductive elimination, which is the proposed pathway.  Precedent for such a pathway is found in platinum(IV) chemistry [9].  The catalytic reaction proceeds at ambient temperature and pressure.  Aluminosilicates have also been used for the ring opening of epoxides [10].

Preliminary Results

            We have begun some initial preliminary research on the catalyzed hydrolysis of carbon-oxygen bonds in ethers [11].  We have investigated three systems.  We find that the molybdate nucleophile PPN₂[MoO₄]

also catalyzes the ring opening hydrolysis of butylene oxide and trimethylene oxide to 1,2-butanediol and 1,3-propanediol respectively, although the yields are low.  These two hydrolyses, in addition to the hydrolysis of tetrahydrofuran to 1,4-butanediol, dioxane to ethylene glycol, and anisole to a mixture of phenol and methanol, also occurs in low yield in a heterogeneous system with aluminum phosphate.  Higher yields for the hydrolysis of anisole are obtained using aluminum trichloride in a Friedel-Craft like reaction.  The reaction conditions of this latter reaction involve hydrolysis of the product obtained between aluminum trichloride and anisole.  Although this reaction is unlikely to develop into a useful one, it does however show that the hydrolysis of anisole is a viable reaction.  The aluminum phosphate reaction may merit further study.  The PPN₂[MoO₄] catalyzed hydrolysis correlates with ring strain whereby the more strained 3- and 4-membered ring ethers undergo hydrolysis, but the 5- and 6-membered rings are unreactive.  Higher nucleophilic strength, or higher electrophilicity of the ether carbon induced by metal ion complexation, may be required to achieve the hydrolytic cleavage of these rings.  Also, a different nucleophile other than molybdate may be necessary.  The PPN salt was used because it gave the molydate ion a higher solubility in an organic medium [8], but the PPN cation cannot bind to the ether oxygen to increase the electrophilicity of the carbons.  Since hydrolytically stable molybdate esters are not available, this is not a plausible route to using the alkali metal salts of molybdates as hydrophobic nucleophiles.

Proposed Research

            This proposed research is targeted toward understanding factors that facilitate the hydrolytic cleavage of carbon-oxygen bonds.  The proposed strategy builds on systems that have been used for the hydrolysis of carbon-oxygen bonds in epoxides to yield alcohols.  Such systems have involved initial nucleophilic attack at the ether carbon, followed by hydrolysis of the alkyl ester bond.  In particular, catalytic systems will be sought that target a synergistic binding of a metal cation to the ether oxygen in order to induce attack by the nucleophilic anion at the ether carbon.

Simple Nucleophiles

            Our initial approach will be to follow a strategy that builds on the catalyzed hydrolysis of ethylene oxide.  An alternative epoxide that can be reacted with water in the presence of a molybdate salt as catalyst is butylene oxide.  Since both this compound and the 1,2-butanediol that is formed (equation 4) are liquids, they will be easier to handle in our laboratory than is ethylene oxide.  Other anionic catalysts such as sodium sulfate, sodium nitrate and sodium phosphate will be used, or the PPN⁺ or

n-butylammonium salts if the biphasic nature of the reaction mixture becomes a problem.  Less reactive carbon-oxygen bonds will be investigated.  Two potential targets are the hydrolysis of tetrahydrofuran and anisole to give 1,4-butanediol, and a mixture of phenol and methanol respectively.  Another target molecule is methyl tert-butyl ether (MTBE).  This compound is chosen because our computational data in Table 1 suggest that the hydrolysis of ethers having tertiary butyl substituents is more favored than the methyl derivative (equation 5).  However, since steric factors may negate this small advantage, MTBE

has the advantage of having both a methyl and a tertiary butyl carbon substituents on the ether, thereby having both steric extremes.  If hydrolysis is observed, isotopic labeling of the water oxygen should allow for identification of the site of attack at the ether carbons.  We anticipate attack at the methyl ether oxygen.

            The mechanism of the hydrolysis of epoxides is proposed to involve the formation of a macrocycle that undergoes subsequent hydrolysis to the diol [8].  An alternate pathway could involve the formation of only one molybdenum alkoxide bond.  Either sequence involves hydrolysis of such bonds.  This hydrolytic cleavage of the bond between the alkyl carbon and the oxygen atom of the nucleophile completes the catalytic cycle.  For many stronger nucleophiles, this hydrolytic cleavage is less facile.

Nucleophile Enhancement

            Another facet of designing catalysts for ether hydrolysis is the nature of the nucleophile.  In one respect, the oxoanions are a good choice because the alkylated nucleophile formed in the first step of the cycle can readily undergo hydrolysis to yield the alcohol and regenerate the oxoanion.  In another respect, however, oxoanions are a poor choice because they are only weak nucleophiles (Table 2), and therefore their reactions

Table 2.           Relative Nucleophilicities of Oxoanions

a. For nucleophilic reactions of para-nitrophenyl acetate in water at 25°C

with carbonium ions are likely to be rather slow [12, 13].  Because of the advantage of ready hydrolysis, we propose to target the enhancement of these nucleophilic reactivities.  One approach to accomplish this is to bind the sodium ion into a complexant, thereby reducing ion pairing with the oxoanion.  From literature precedent this can be achieved by the addition of a crown ether that has the correct cavity size to strongly bind the sodium cation [10].  Another approach is to use a calixarene complexant.  Calixarenes have several advantages in that within the same molecule they can have separate oxoanion and sodium ion binding sites, and also they are good phase transfer catalysts.  The proposed catalysts in this family of compounds is 1, which can be prepared in good yield using the sequence of reactions in scheme C.  In this compound the methoxy phenyl ether site is a complexant for the sodium ion, leaving the phosphonate site as a deprotonated anion.

Phase Transfer Aspects

            Other approaches to designing catalytically active nucleophiles will be followed.  These involve making modifications that will influence both the nucleophilicity and the phase transfer properties of the oxoanion.  We plan to focus these efforts on molybdate, sulfonates and phosphonates.  For molybdate, we will focus on the phase transfer aspects because introducing substituents that are stable to hydrolytic cleavage on such inorganic oxoanions is not a realistic option.  Two media that may allow sodium molybdate itself (rather than its PPN salt) to be used as an effective catalyst for carbon-oxygen bond cleavage are poly(ethylene glycol) and fluorocarbons.  These media both show biphasic behavior.  Interconversion between two phases and a single phase can be accomplished with poly(ethylene glycol) by the addition of salts [14-18], and with fluorous phase compounds by changes in temperature [19-21].  Another alternative is to use long chain alkylammonium cations such as the butyl, octyl, dodecyl, or hexadecyl derivative to increase the hydrophobicity of the molybdate salt.  Similar considerations would

apply to other oxoanions such as tungstate, vanadate, or nitrate. For sulfonates and phosphonates, however, substituents can be introduced to change their character.  The simplest way to accomplish this is to introduce long chain alkyl groups at the sulfur and phosphorus centers.  Such compounds having both a hydrophobic and a hydrophilic end are nucleophiles that can also act as phase transfer catalysts.  Examples of two such compounds that will be investigated are 2 and 3.  Each of these compounds with the heteroatom in its highest oxidation state is oxidatively stable.  One difference, however, is that the alkylsulfonic acid 2 is a stronger protonic acid than is the alkylphosphonic acid 3, therefore 2 will be dianionic in solutions of higher acidities.  Although our initial choice of substituents will be the n-octyl, the desired phase transfer properties may required a longer alkyl chain substituents such as the n-octadecyl (n-C₁₈H₃₇) derivative.  This choice will need to be made by experiment.

            Cooperativity

            Another approach will be to try and induce cooperativity between the oxoanion nucleophile and its cation in the salt.  If the cation can be complexed to the ether oxygen, it will increase the electrophilicity of the carbon atom covalently bound to this oxygen.  The nucleophile system used in scheme B has the disadvantage that by using the PPN⁺ salt to achieve solubility, there is no possibility of cooperative binding of the cation to the ether oxygen.  Sodium is a good choice for the cation because it can bind to the ether oxygen.  Phase compatibility must then be achieved in a different manner, and since water is present in the catalytic system, it will be competitive with the ether oxygen for binding to the cation.  Lithium and magnesium are alternate choices because both have higher cationic charge densities than does sodium.  Higher charge densities will result in greater induced electrophilicities at the alpha carbon of the ether if these ions bind to the oxygen.  A limitation, however, is that salts of the higher charged cations are more likely to be insoluble.  Our choice of cation will be a combination of solubility properties and the function of the cation to induce electrophilicity at the ether carbon.

Organometallic Approach

            Another approach to inducing the cleavage of a carbon-oxygen bonds in an ether is to use an organometallic strategy.  Among the possible nucleophiles that may be methylated by anisole are the platinum(0) complexes Pt(PEt₃)₃ and Pt(PCy₃)₂, and the sodium salts of cpMo(CO)₂(PPh₃)^-.  These organometallic salts have the advantage that cooperativity is possible between the sodium cation and the organometallic anion.  Identification of the methylmetal complex that is formed can be readily achieved by a combination of hydrogen-1 and carbon-13 NMR spectroscopy.  A disadvantage of using an organometallic approach is the difficulty of achieving a catalytic system, but it is useful to investigate it as a stoichiometric reagent because it will provide information about the reactivity that is required to induce ethers to react as electrophilic alkylating agents to metal centers.

Reaction Pathways

            Although we have assumed that these cleavage reactions will occur by a simple nucleophilic displacement reaction, there are other possible pathways.  Another alternative is an electron transfer pathway.  Such a pathway may be the predominant one when the nucleophile is sterically blocked from directly approaching the electrophilic carbon on the ether.  An electron transfer pathway leading to the nucleophile induced cleavage of a carbon-oxygen bond in an ether is shown in Scheme 2.  Again, the alcohol ROH is the hydrolysis product.

Nuc^-                 +                      ROR                ®                    Nuc                  +          ROR^-

                                                ROR^-                ®                    RO^-                  +          R

RO^-                  +                      H₂O                 ®                    ROH                +          HO^-

Nuc                  +                      R                      ®                    NucR

NucR               +                      H₂O                 ®                    NucH               +          ROH

NucH               +                      Base                 ®                    Nuc^-                 +          BaseH⁺

Scheme 2

Ammonolysis

            Although the catalyzed hydrolysis of the carbon-oxygen bond is the major focus of this proposed research, we also plan to investigate whether ammonolysis can  be achieved (Table 3).  Similar reactions

Table 3.           Computational Enthalpies (kcal mol ^-1) for the Ammonolysis of Ethers

(ΔH₁, semi empirical (PM3) methods.  ΔH₂, density functional theory)

a.  Ammonolysis to PhOH and MeNH₂  b.  Ammonolysis to PhNH₂ and MeOH

conditions to the hydrolysis studies will be used, except that aqueous ammonia will be used in place of water (equation 6).  The higher nucleophilicity of ammonia over water may lead to ammonolysis being competitive with hydrolysis.

Thioethers

            A second hydrolysis and ammonolysis system that will be investigated is thioethers.  Our computational enthalpy data show that the hydrolysis of sulfides is either enthalpically favorable, or that the ones that are endothermic can possibly be made thermodynamically favorable by the entropic term (Table 4). 

Table 4.           Computational Enthalpies (kcal mol ^-1) by Density Functional Theory for the Hydrolysis of Thioethers

The targeted hydrolysis reactions is shown in equation 7.  In targeting thioether hydrolysis a slight variation to our strategy will be made.  This variation will be in the cation activation step.  For the cationic activation of ethers to nucleophilic attack at carbon we proposed using cations such as sodium,                      

lithium, and magnesium.  These are “hard” cations that bind to oxygen donors.  For thioethers we need a cation that will bind to the sulfur center.  Since sulfur is a “soft” donor, we will need to use a “soft” cation to selectively bind to it.  Of the soft cations that we propose to use as cations are lead(II) and cadmium(II), although we would prefer a less toxic metal such as palladium(II).

Analytical Methods

            For this research project we will analyze our hydrolysis and ammonolysis products by a combination of chromatography (thin layer, gas, and liquid) and spectroscopy (IR, NMR, and mass), against authentic samples if possible.  If we are successful in achieving catalytic carbon-oxygen bond catalysis we will try our system on biopolymers such as starch, humus material, and lignin, and we will also investigate methods to prepare solid state materials that have the same or better properties.

References

[1]        Platinum Metal Complexes of Amine and Ether Substituted Phosphines, T. B. Rauchfuss, F. T. Patino and D. M. Roundhill, Inorg. Chem., 14, 652 (1975).

[2]        Kinetics and Mechanism of the Reaction of Palladium(II) Complexes of o_Diphenylphosphino-thioanisole and o_Diphenylphosphinoselenoanisole with the Nucleophiles Thiocyanate and Iodide, D. M. Roundhill, W. B. Beaulieu and U. Bagchi, J. Am. Chem. Soc., 101, 5428 (1979).

[3]        Menschutkin Type Amine Alkylations Involving Ethyl Transfer from Platinum(II) Chelate Complexes of o_(Diphenylphosphino)thiophenetole, A. Benefiel and D. M. Roundhill, Inorg. Chem., 25, 4027 (1986).

[4]        Characteristics of Nickel (0), Nickel (I) and Nickel (II) in Phosphino Thioether Complexes: Molecular Structure and S-Dealkylation of (Ph₂P(o-C₆H₄)SCH₃)₂Ni^o, J.S. Kim, J.H. Reibenspies and M.Y. Darensbourg, J. Am. Chem. Soc., 118, 4115 (1996).

[5]        Synthesis and Structural Investigation of Polyoxomolybdate Coordination Compounds Displaying a Tetranuclear Core.  Crystal and Molecular Structures of [n-Bu₄N]₂[Mo₄O₁₀(OMe)₄X₂] (X = -OMe, -Cl) and Their Relationship to the Catecholate Derivative [n-Bu₄N]₂[Mo₄O₁₀(OMe)₂(OC₆H₄O)₂] and to the Diazenido Complexes of the o-Aminophenolate and the Naphthalene-2,3-diolate Derivatives [n-Bu₄N]₂[Mo₄O₆(OMe)₂(HNC₆H₄O)₂(NNC₆H₅)₄] and [n-Bu₄N]₂[Mo₄O₆(OMe)₂(C₁₀H₆O₂)₂(NNC₆H₅)₄]. Comparison to the Structure of a Binuclear Complex with the [Mo₂(OMe)₂(NNC₆H₅)₄]²⁺ Core, {Mo₂(OMe)₂(H₂NC₆H₄O)₂(NNC₆H₅)₄, H. Kang, S. Liu, S.N. Shaikh, T. Nicholson, and J. Zubieta, Inorg. Chem. 28, 920 (1989).

[6]        Hexanuclear Complexes of Molybdenum(V) containing [Mo₆O₁₂(OCH₃)₄(acac)₃]- anion, M. Cindric´, G. Pavlovic´, V. Vrdoljak, B. Kamenar, Polyhedron, 19, 1471 (2000).

[7]        Fecp⁺ Induced Heterolytic Cleavage of Phenyl Ethers and Transetherification, L. Djakovitch, F. Moulines and D. Astruc, New J. Chem., 20, 1071 (1996).

[8]        Selective Ethylene Oxide Hydrolysis Catalysed by Oxo-Molybdenum Species, J. R. Briggs, A. M. Harrison and J. H. Robson, Polyhedron, 5, 281 (1986).

[9]        Direct Observation of C-O Reductive Elimination from Pt(IV), B. S. Williams, A. W. Holland and K. I. Goldberg, J. Am. Chem. Soc., 121, 252 (1999).

[10]      Organic Syntheses using Aluminosilicates, Y. Izumi and M. Onaka, Adv. In Catal. 38, 245 (1992).

[11]      Nucleophile assisted hydrolysis of carbon-oxygen bonds in ethers, C. Polydore, D.M. Roundhill, H.-Q. Liu, J. Mol. Catal. A: Chemical, 186, 65 (2002).

[12]      General Base and Nucleophilic Catalysis of Ester Hydrolysis and Related Reactions, S. L. Johnson, Adv. Phys. Org. Chem., 5, 237 (1967).

[13]      The Mechanisms of Reactions of ß-Lactam Antibiotics, M. I. Page, Adv. Phys. Org. Chem., 23, 165 (1987).

[14]      Manipulation of Nucleophilic Displacement Reactions by Host-Guest Complexes, H.-J. Scheider, R. Busch, R. Kramer, U. Schneider, I. Theis, Adv. in Chem. Ser., 215, 457 (1987).

[15]      Poly(ethylene Glycol) Chemistry: Biotechnical and Biomedical Applications, J. M. Harris, ed., Plenum, New York, 1992.

[16]      Aqueous Biphasic Separations: Biomolecules to Metal Ions, R. D. Rogers, A. H. Bond, C. B. Bauer, J. Zhang, M. L. Jezl, D. M. Roden, S. D. Rein and R. R. Chomko, Plenum, New York, R. D. Rogers and M. A. Eiteman, eds., 1995, pp. 1-20.

[17]      Ion Exchange and Solvent Extraction, R. D. Rogers and J. Zhang, Dekker, New York, J. A. Marinsky and Y. Marcus, eds., 1997, Vol. 13, Chapter 4, pp. 141-193.

[18]      Metal Ion Separations in Polyethylene Glycol-Based Aqueous Biphasic Systems, R. D. Rogers, A. H. Bond and C. B. Bauer, Sep. Sci. Technol., 28, 1091 (1993).

[19]      Synthesis and Characterization of Calix[4]arene Functionalized Poly(ethylene glycol) Derivatives, J. Shen, H. F. Koch and D. M. Roundhill, J. Incl. Phenom. Macrocycl. Chem. 38, 57 (2000).

[20]      Facile catalyst Separation without Water: Fluorous Biphase Hydroformylation of Olefins, I. T Horváth and J. Rábai, Science, 266, 72 (1994).

[21]      Fluorous Biphase Chemistry, I. T Horváth, Accts. Chem. Res., 31, 641 (1998).

[22]      Molecular Engineering in Homogeneous Catalysis: One-Phase Catalysis Coupled with Biphase Catalyst Separation.  The Fluorous-Soluble HRh(CO){P[CH₂CH₂(CF₂)₅CF₃]₃}₃ Hydroformylation System, J. Am. Chem. Soc., 120, 3133 (1998).

1.         C. D. Gutsche, Calixarenes, Royal Society of Chemistry: Cambridge, 1989.

2.         C. D. Gutsche, Calixarenes Revisited, Royal Society of Chemistry: Cambridge, 1998.

3.         J. Vicens, V. Böhmer, Eds., Calixarenes: A Versatile Class of Macrocyclic Compounds, Kluwer Academic, Dordrecht, 1991.

4.         J. Vicens, Z. Asfari, J. B. Harrowfield, Eds.; Calixarenes: 50^th Anniversary, Kluwer Academic, Dordrecht, 1994.

5.         J. Vicens, V. Böhmer, Calixarenes: A Versatile Class of Macrocyclic Compounds, Kluwer, Dordrecht, Netherlands, 1990.

6.         Z. Asfari, V. Böhmer, J. B. Harrowfield,  J. Vicens, Eds.; Calixarenes 2001, Kluwer Academic, Dordrecht, 2001.

7.         G. J. Lumetta, R. D. Rogers, A. S. Gopalan, Eds., Calixarenes for Separations, ACS Sympos. Series 757; American Chemical Society: Washington, DC, 2000.

8.         L. G. Sillen, A. E. Martell, Stability constants, Special Publication No. 17, The Chemical Soceity, London, 1964.

9.         M. J. Taylor, Comprehensive Coordination Chemistry, G. Wilkinson, ed., Pergamon, 1987, Vol.   3, Chapter 25.1.  

10.        D. M. Roundhill, Metal Complexes of Calixarenes, Prog. Inorg. Chem., 1995, 43, 533.

11.        D. M. Roundhill, Extraction of Metals from Soils and Waters, Plenum, New York, 2001.

            e)         Molecular mechanics calculations of ligands and complexes.

            a)         Hexadentate Complexes of Gallium(III), Indium(III) and Iron(III)

            An alternate strategy for the complexation of gallium(III), indium(III) and iron(III) involves the use of multidentate ligands.  For a high solution stability of the complex, a hexadentate ligand is expected to be particularly advantageous.  Previous work on complexes of gallium(III) and indium(III) has shown that phenolate ions bind strongly to these metal ions, therefore we have synthesized a hexadentate chelate ligand incorporating such a functionality.  Previous work has suggested that such ligands that contain a carbon-nitrogen double bond undergo in vivo degradation, therefore we have avoided such functionalities in the ligands by carrying out a final reduction step with KBH₄.  The ligands have been synthesized in 30-40% yield by the synthetic procedure outlined in scheme 2. These ligands

Scheme 2

(XsalH₂tach (X = H, NO₂, OMe) have been fully characterized. (Bollinger, 1995).  The uncharged 1:1 complexes between these ligands and trivalent aluminum,  gallium, indium and iron have been synthesized.  The aluminum, gallium and indium complexes have been characterized by a combination of IR, UV and ¹H and ¹³C{¹H} NMR spectroscopy.  The iron complexes have been characterized by a combination of IR and UV spectroscopy and magnetic measurements.  Both sets of complexes have been characterized by single crystal X-ray crystallography.  The complexes have sufficient solubility in water, for in vivo imaging applications.  The complexes are stable both in the solution and solid state.  The gallium complex in solution in the presence of excess iron(III) shows no replacement of gallium(III) after 1 week, as evidenced by the lack of any red coloration due to the formation of the iron(III) complex.

            This 1, 3, 5- triaminocyclohexane backbone has been used to develop analogous ligands that are used to bind metal ions (Park, 2000). Among the metal ions that have been studied are gallium(III), indium(III), and iron(III)  (Hilfiker, 1997; Park, 1998; Torti, 1998). These compounds are similar to other multidentate systems (Setyawati, 1999; Orvig, 1999), except that the conformation of the cyclohexane ring sets up the preferred coordination geometry for complexation.

            b)         In Vivo Aluminum and Iron Chelate Studies

            In vivo studies have been carried out in collaboration with Dr. W. Banks at the Veterans Administration Hospital in New Orleans.  The three ligands have been used to investigate their effect on the transfer of aluminum and iron across the blood-brain barrier (Bollinger, 1995).The procedures used for the in vivo animal studies with these new compounds follow closely those that have been used previously in these laboratories and described elsewhere (Banks, 1988).  The mice used in this study were male ICR mice purchased from Charles River Laboratories, Wilmington, MA.  All mice weighed 17-20g.  To preserve consistency, the given procedures were conducted simultaneously in three groups as a single experiment.  The three groups, saline control, aluminum control and aluminum/chelator, consisted of 6-7 mice apiece.  Experiments were repeated three to four times for a total of about 20 mice per group.  Mice in the experimental group and the aluminum control group received an i.p. injection of aluminum (100 mg/kg) in normal saline (0.2 mL).  The saline control group received an acidified saline solution (0.2 mL).  Immediately afterwards, mice in all groups received a s.c. injection over the thoracic spine.  The experimental group received a sample of the ligand dissolved in the solution containing 5% DMSO in the saline control solution (0.2 mL of 0.6 mM) (solution 4).  The other two groups received only the pure solution of 5% DMSO in the saline solution (0.2 mL). 

            After about 35 min, the mice were anesthetized with urethane (40% in saline, 0.2 mL) by i.p. injection.  The scalp was then exposed and a hole 3.0 to 3.5 mm deep was made into the left lateral ventricle of the brain 1.0 mm lateral and 1.0 mm posterior to the bregma.  This was accomplished with a 26-gauge needle sheathed with polyethylene tubing to cover all but the terminal 3.5 mm (Noble, 1967).  After a maximum of 65 min following the first i.p. and s.c. injections, an intracerebroventricular injection (i.c.v.) of 1 mL lactated Ringer's solution containing 25 x 10³ cpm radioactive peptide (¹²⁵I labeled Tyr-MIF-1) into this ventricle was made using a 1 mL Hamiltonian syringe (Hamilton Co., Reno, NV).

            Mice were decapitated 20 min after i.c.v. injection and their brains removed, excluding the pineal and pituitary.  The brains were placed individually into 10 x 75 mm test tubes and centrifuged for 5 min at 2500 g to seat them at the base of the tubes.  Each brain was then counted for 3 min with a gamma-counter (Micromedic 10/200, Horsham, PA) to determine the level of residual, or non-transported, peptide in the brain.  Groups were compared for statistically significant differences by analysis of variance (ANOVA) followed by Duncan's multiple range test.

            The effect of aluminum on the transport system over the 20 min time period was determined by the inclusion of a 0.2 mL i.p. injection of 89 mg/mL aluminum chloride hexahydrate in normal saline. It has been previously shown that the most dramatic effect of aluminum upon the transport of Tyr-MIF-1 occurs 60 - 90 minutes after i.p. injection in experiments involving a 10 minute incubation period (Banks, 1989). This general time frame was adhered to for the present experiments, such that the entire 20 min incubation period fell between 60 and 90 min after the injection of aluminum. After analysis of the brains from control mice, it was found that the amount of  peptide not transported averaged about 5500 cpm/brain.  After treatment with aluminum, this increased to about 7400 cpm/brain.  In all cases, this represented a statistically significant increase. However, this increase is less than that typically seen and suggests that inhibition of PTS-1 by aluminum in the current study was less than maximal. This may have been due the presence of DMSO, injected s.c. in all groups as a diluent, which may have sequestered some of the aluminum from the circulation.

            These results suggest that the (NO₂salH₂)₃tachH₃ chelator is capable of binding aluminum from the circulation and that the resulting complex may enhance delivery of biologically active aluminum to the site of action. Because it is not known whether the effect of aluminum occurs in the blood or brain side of the BBB, speculation on the mechanism of its enhancement by this ligand is difficult. The nature of the interaction of metal and (NO₂salH₂)₃tachH₃ in vivo is also speculative. The active (NO₂salH₂)₃tachH₃ aluminum complex exhibits much higher octanol/water partition coefficients than those for the other two complexes with the inactive ligands, as well as much lower solubility. It is possible, as a result, that this ligand aids in the delivery of aluminum through or into the membranes composing the BBB and does so more efficiently or rapidly than its less lipophilic analogs. This model requires that the complex itself is biologically active or that after such delivery, the aluminum is released into its active form.

            Similar N₄O₃ tripodal amine phenols have been used to synthesize complexes with Al³⁺, but in this case they are cationic (Liu, 1992). Formation of similar charged species other than simple 1:1 complexes cannot be ruled out in these experiments.  Because any discrete multi-ligated metal complex would have a resultant charge, these species would not be expected to readily cross membrane barriers. They might, in some way, enhance the presence of the active aluminum at the site of action, and might also act to prevent clearance of aluminum by the kidney from the bloodstream, effectively enhancing blood levels of the metal. It is unlikely that the nitrated ligand, although observed to form apparent crosslinked gels in vitro, is combining with aluminum to form a physical barrier to peptide migration.  Thus, the results show that (NO₂sal)₃tachH₃ significantly affects the action of aluminum on the PTS-1 transport system.  In addition these ligands have been studied as for their ability to transfer radioactive iron into selected regions of the body.  Use of these ligands and their complexes as imaging agents and iron-sequestering agents is presently under investigation in vivo.  These experiments use radioactive ⁵⁹Fe(III) in mice as a model for both indium and gallium radioimaging agents, as well as for the use of the free ligands as scavengers of iron.  Preliminary results show that the radioactive complexes of each of the three ligands are cleared rapidly from the bloodstream as compared to uncomplexed Fe(III), with those of the methoxy ligand cleared fastest, followed by the benzyl and nitro substituted ligands, respectively.  These results suggest that if the free ligands are capable of penetrating tissues and then coordinating iron, the resultant complexes could then be removed from serum.  Despite rapid clearance, the complexes do show a different pattern of selective tissue uptake when compared to transferrin-bound iron, especially that of the nitro ligand.  Over a two hour time period, this complex appears to enhance uptake into lung, skin, adrenal, liver and kidney tissues, while diminishing spleen, bone and brain uptake.  The other two complexes show evidence of rapid association with brain and muscle tissue.  Over longer periods, differences of uptake in various tissues begins to match more closely the uptake of the non-complexed iron control group, implying that metabolism of the complexes releases free iron.  Despite this metabolism, some differences in experimental and control groups remain among the ligands in certain tissues for at least 24 h.

            c)         Synthesis of  2-Hydroxyethoxy, 2-Amidoethoxy and 2-Aminoethoxy           Substituted Calixarenes

            We have recently accomplished the synthesis and characterization of a series calix[4]arenes, calix[6]arenes and calix[8]arenes having alcohol, amine, or amide functionalities appended onto the lower rim, and either tert-butyl groups or hydrogens attached to the para positions of the upper rim (Moran, 1992, 1995).  We have used two approaches to changing the number of alcohol groups on this rim.  The first method involves blocking several of the hydroxylated sites on the lower rim with methoxyl groups, and the second one involves changing the number of hydroxy groups on the rim by using either a calix[4]arene, a calix[6]arene or a calix[8]arene.

            The methoxylated calixarenes have been converted by treating the calix[6]arene with methyl iodide or methyl tosylate (MeX) (equation 1).

(1)

The amide derivatives have been prepared by treating the calixarene with either bromoacetamide or a substituted bromoacetamide (equation 2)

(2)

Reduction of these calixarene amides with hydride reagents gives the amines in good yield (equation 3).

(3)

An alternate synthesis from chloroacetonitrile has also been developed using the procedure shown in equation 4 (Roundhill, 1994).

(4)

4.         Research Design and Methods

            In this proposal we describe a strategy to design a new class of encapsulating ligands for metal ions based on calixarenes.  The proposal is primarily synthetic, but we plan to use molecular mechanics calculations and X-ray crystallography to design ligands and judge which particular compounds appear to have the most optimal combination of distances and geometry for binding trivalent metal ions.

            The decision to use calixarenes as ligands is based on a number of factors.  The most important of these is that the calixarenes have two distinctly different rims onto which substituents can be separately appended.  One rim can be used for the specific purpose of incorporating functional groups that will bind the metal ion, and the other rim can be used to incorporate groups that will either impart the desired solubility characteristics to the complex, or will provide the necessary functional groups for the chemical attachment to carrier proteins.  A further reason for choosing the calixarenes is that they are conformationally mobile, and that they have a variable number of binding sites available for functionalization.  The combination of these two features results in the compounds offering a broad-range of options for the design and synthesis of ligands for imaging purpose (Bryant, 2000).

            Although this proposal is focused on a select number of metal ions, the proposed ligands are not necessarily restricted to these.  Indeed, a strength of this calixarene approach is that it eliminates the need to carry out extensive synthetic studies to build a ligand platform for each metal ion to determine the optimal ligand for use.  Even though some modification of the calixarene core will likely be necessary to accommodate metal ions of different sizes, the basic concept should be applicable to any particular metal ion.  The concept is analogous to that being used by Cram and others for purely organic host-guest compounds, where the host molecule has been termed a carcerand, and the guest molecule is trapped inside (Cram, 1991, 1992, 1993).  Although similar strategies for metal ion complexation have been previously used (Boston, 1968, Busch, 1993) with more conventional multidentate chelate ligands, we believe that the synthetic opportunities available with the calixarenes to both control the solubility characteristics and to append a single functional group to bind to a protein or antibody carrier makes them an attractive alternative to the ligand types presently being used and studied.  This attractiveness of calixarenes comes from the possibility that we can make a single, or a small number of ligands, that will bind a wide variety of metal ions for diagnostic or therapeutic applications.  The elegance of our approach is that we do not need to design a new specific ligand with a high stability constant for each metal ion.   Instead the encapsulated metal ion has a reasonably good chance of remaining bound unless the in vivo environmental causes extensive bond breaking to occur leading to degradation of the cage, and release of the metal ion.  The calixarenes are particularly attractive because chemical groups can be separately appended to each rim.  The functional groups on one rim are available for incorporating the encapsulating ligand, and those on the other rim are available for modifying the solubility characteristics or the functional groups that will be used for chemical attachment to an antibody or carrier protein.

            This proposal develops from our recent focus on the (RsalH₂)₃tach ligands.  Although this research has led to the synthesis of new lipophilic complexes that interact with the blood-brain barrier (Bollinger, 1995), their architecture makes them difficult to modify for changing the solubility characteristics of the complexes or for their attachment to polymers by single strand attachment.  This proposal is therefore focused on synthesizing a metal ion coordination site like that of (RsalH₃)₃tach, but since it is assembled from a calixarene, we will have the chemical advantages of this host to incorporate the desired ligand properties.                                                                                                        

            a)         Synthesis of Narrow Rim Modified Calixarenes

            We propose to use calixarenes with functional groups appended to either an azacalix[3]arene or the narrow rim of a calix[6]arene as precursor compounds for the synthesis of encapsulating ligands for gallium(III), indium(III) and yttrium(III) ions.  An advantage of using narrow rim modified calixarenes is that this rim has oxygens appended, and the particular metal ions that are targeted have a preference for such oxophilic ligands.  We propose to use mixed functionality calixarenes with both phenolic groups and amine groups appended.  Two strategies will be followed for using these calixarenes in the synthesis of encapsulating ligands.  The first strategy will involve incorporating three amino arms onto an azacalix[3]arene such that the alkoxide groups are still available for deprotonation and complexation.  The second strategy will involve using a reaction with chloroacetonitrile followed by reduction of the nitrile functionality to incorporate amine groups onto the narrow rim of a calix[6]arene. We therefore propose to synthesize these two types of mixed functionality calixarenes so that can be used to prepare uncharged hexadentate complexes with gallium(III), indium(III) and yttrium(III).

            We propose to synthesize N₃O₃ hexadentate chelates with both azacalix[3]arenes and calix[6]arenes that are analogous to those previously prepared by us from cyclohexane-1,3,5-triamine.  Since calixarenes have phenolic groups already available on the lower rim, our synthetic strategy involves retaining three of these functional groups unsubstituted in order to provide the three phenolate ligating groups. Our choice of the azacalix[3]arene platform is based on the availability of three phenolic functionalities that can be deprotonated to give a trianionic ligand, along with three amine sites that can be potentially functionalized with amino groups to give the targeted hexadentate N₃O₃ ligand system 1. Our proposed route to synthesizing such ligands is shown in scheme 3. This strategy involves introducing the amino arms of the hexadentate chelate into the azacalix[3]arene in the synthetic step

rather than in attempting to introduce the substituent onto previously synthesized azacalix[3]arenes. We believe that this approach has the highest likelihood of being successful, and will eliminate the possibility of obtaining mixtures containing azacalix[3]arenes having different degrees of substitution.

            The other group of calixarenes of choice will be calix[6]arenes where the other three phenolic groups can be chemically modified to introduce functionalities that can be capped to obtain the desired calixspherands.  The functionalities of choice for carrying out these capping reactions are alcohols or amines.  From our preliminary work with lower rim substituted calixarenes we plan to introduce 2-aminoethoxy substituents onto alternating positions of the narrow rim.  Previous work by us and others (Moran, 1994; Casnati, 1991) has shown that by controlling the stoichiometries of the reagents it is possible to selectively substitute three of the six phenolic hydrogens with organic groups in high yield.  The proposed synthetic route to the target compound 2 is shown in scheme 4. This selective substitution of three of the six lower rim  phenolic groups of a calix[6]arene can be achieved by controlling the reaction conditions.  The reagents ClCH₂CN and the calix[6]arene will be used in a stoichiometry that is in a 3 : 1 ratio with respect to the calix[6]arene, and the optimal conditions required to obtain the desired

compounds will be determined by liquid chromatographic analysis. Once these experiments are complete the products will be prepared in larger quantities and purified by chromatography.  In the trisubstituted product it is likely that the isomer obtained has the substituents in alternating positions around the lower rim.  This substitution pattern is favored because the ether linkages that are formed in the alkylation reaction are sites for forming hydrogen bonds to the two adjacent phenolic groups.  These hydrogen bonded phenol hydrogens are blocked from reaction with the alkylating agent, thereby causing 1,3- and 1, 3, 5-substitution patterns.  The reduction step in scheme 4 will be carried out with either borane (BH₃) or lithium aluminum hydride (LiAlH₄).  Similar reductions have been successfully used by us in the synthesis of the fully substituted 2-hydroxyethoxy and 2-aminoethoxy calix[6]arenes (Roundhill, 1994), therefore these reagents will be those of choice for the transformations in this scheme.

b)         Synthesis of Neutral Complexes of Gallium(III), Indium(III) and Yttrium(III) with Azacalix[3]arenes and Calix[6]arenes

            Complexation studies between these new 2-aminoethoxy substituted azacalix[3]arenes and calix[6]arenes and gallium(III), indium(III) and yttrium(III) will be carried out by the addition of solutions of the metal salts to solutions of the calixarene ligand.  Water, alcohol or acetonitrile will be the solvents of choice for these reactions, and chloride or nitrate will be the ones of choice for the metal salts.  After the metal and ligand solutions have been either allowed to stand at ambient temperature or refluxed in a high boiling point solvent, an aliquot of the solution will be removed, and its UV-visible or ¹H NMR spectrum measured.  If the solutions show additional absorption bands or band shifts diagnostic of complexation, the complex will be isolated in the solid state.  The isolation procedure will be accomplished either by the addition of either a second solvent or a bulky anion that results in precipitation of the complex.  Possible bulky anions are BF, PF, BPh.  An infrared spectrum of the solid complex will be measured, and band shifts will be sought that are diagnostic of coordination of the gallium(III), indium(III) or yttrium(III) ion to the calixarene.  Possible diagnostic shifts are changes in n(OH) and n(NH) for the alcohol and amine functionalized ligands respectively, or the appearance of new bands in the far infrared spectra due to n(MO) and n(MN).  The most definitive proof of the stoichiometries and structures of the complexes that are formed will be by single crystal X-ray crystallography.  Single crystals of the complexes can be grown from solution by a combination of techniques that include solvent layering and slow diffusion.  Previously we have collaborated with Dr. Arnold Rheingold at Delaware and Dr. Frank Fronczek at Louisiana State University for X-ray crystallography, and we will do so again if crytals of the complexes are obtained.

            Where suitable crystals of the isolated complex cannot be grown, the characterization will be carried out by a combination of ¹H and ¹³C{¹H} NMR, IR, UV and mass spectroscopic techniques.  In order to obtain a mass spectrum of these complexes it will be necessary to use fast atom bombardment (FAB) methods because the complexes will have too a low volatility for electron impact methods.  Coordination of the metal ion to the 2-aminoethyl substituted calixarene is expected to lead to shifts in the ¹H and ¹³C{¹H} NMR spectra in the complexed ligand.  If the metal ion is coordinated to both the phenoxide and amine groups at the calixarene periphery we expect that there will be shifts in the resonances of the ligand hydrogen and carbon atoms that are in close proximity to the metal binding site.

            The complexes that are targeted are those with the tris-2-aminoethyl calixarenes 1 and 2 shown in schemes 3 and 4.  These choices of 1 and 2 are made because we can potentially synthesize uncharged complexes of the trivalent gallium(III), indium(III), and yttrium(III) with them.  Uncharged complexes have the advantage of being lipophilic, and since the complexes with 1 will have a low molecular weight, they have the potential to be permeable to the blood brain barrier.  The targeted complexes are expected to be uncharged with these trivalent metal ions because coordination should occur with the metal covalently bonded to three phenoxide groups.  There is a strong likelihood that this selectivity of complexation will be observed.  The phenolic groups in these chosen calixarenes are the functionalities within the ligands that have the highest acidities, and it is documented that gallium(III), indium(III) and yttrium(III) form strong complexes with phenoxides (Carty, 1975; Tuck, 1983; Motekaitis, 1990).  Furthermore, these ligands 1 and 2 are closely analogous to our (XsalH₂)₃tach ligands that gave neutral hexacoordinate M(XsalH₂)₃tach (M = Ga, In, Fe) complexes.  In these complexes coordination occurs via a combination of three phenoxide likely that the isomer obtained has the substituents in alternating positions around the lower ions and three amine nitrogens.

            The lipophilicities of the complexes of gallium(III), indium(III), and yttrium(III) with 1 and 2 will be estimated by measuring their partition coefficients between 1-octanol and water.  High lipophilicity is potentially advantageous in obtaining complexes that can pass through cell membranes or through the blood brain barrier, therefore it is a useful property to target.  Our previously prepared complexes of gallium(III), indium(III), and aluminum(III) with the deprotonated  (XsalH₂tach (X = H, NO₂, OMe) series of ligands show high lipophilicities, therefore there is a good likelihood that these analogous calixarene complexes will be similar in this respect.  The stabilities of the complexes toward displacement of gallium(III), indium(III), or yttrium(III) by iron(III) (a common metal ion in an in vivo environment) will be estimated by adding aliquots of iron(III) to solutions of the complexes of these three, and determining if metal ion displacement occurs.  One possible way of monitoring such a displacement reaction is to determine if any red coloration is observed that is diagnostic of the formation of iron phenolates (Bollinger, 1995).

c) Binding constants

            The binding constants of the hexadentate azacalix[3]arene and calix[6]arene products 1 and 2 obtained in schemes 3 and 4 with gallium(III), indium(III), yttrium(III), and, for comparison, iron(III), will be measured. These measurements will be made using either potentiometry or UV- visible spectroscopy (Sillen, 1964). Either or both methods will be used depending on whether a chromophore change is observed upon complexation or whether the system has sufficient solubility in water that a pH electrode can be used. We anticipate that the presence of multiple amine functionalities appended to these calixarenes should lead to them having sufficient aqueous solubility that these binding constant measurements can be made in this medium.

            These binding constants will allow us to access the potential of these compounds for imaging or therapeutical applications. The goal is to find the combination of functional groups that lead to high binding constants for gallium(III), indium(III) or yttrium(III), but lower binding constants for iron(III). This property will allow for the metal complexes to retain their integrity in the in vivo presence of iron(III). These comparative binding constant data for these two groups of calixarenes will allow us to decide whether the shape and size characteristics of the calix[3]arene or the calix[6]arene platform are preferable for designing ligands for gallium(III), indium(III), or yttrium(III). At that stage of the research we will seek a collaborator to carry out the biological studies with us. In the past we have collaborated with both Dr. Michael Welch at the Medical School at Washington University in St. Louis, and Dr. William Banks at the Tulane University School of Medicine.  Either would likely be willing to collaborate with us again, but this project is not yet at the stage where such a collaboration is necessary for completion of the proposed research.

 Literature Cited

Alavi, A.; Hirsch, L. J., "Studies of central nervous system disorders with single photon emission computed tomography and positron emission tomography: evolution over the past 2 decades", Semin. Nucl. Med., 1991, 21, 58.

Alvarez, V. L.; Wen, M. L.; Lee, C. et. al., "Site-specifically modified indium-111-labeled antibodies give low liver backgrounds and improved radioimmunoscintigraphy", Nucl. Med. Biol.,  1986, 13, 347.

Anderson, C. J.; Welch, M. J., “Radiometal-Labeled Agents (Non-Technetium) for Diagnostic Imaging”,

            Chemical Reviews (Washington, D. C.), 1999, 99, 2219.

Banks, W. A.; Kastin, A., "Quantifying carrier-mediated transport of peptides from the brain to the blood", in: Hormone Action, Part K, Neuroendocrine Peptides, P. Michael Conn (ed.), Meth. Enzymol., 1989, 168, 652.

Banks, W. A.; Kastin, A. J.; Fasold, M. B., "Differential effect of aluminum on the blood-brain barrier transport of peptides, technetium and albumin", J. Pharmacol. Exp. Ther., 1988, 244, 579.

Becker, W. S.; Behr, T. M.; Cumme, F.; Roessler, W.; Wendler, J.; Kern, P. M.; Gramatzki, M.; Kalden, J. R.; Goldenberg, D. M.; Wolf, F. G.,  ^“67Ga citrate versus  ⁹⁹mTc-labeled LL2-Fab’ (anti-CD22) fragments in the staging of B-cell non-Hodgkin’s lyphoma”, Cancer Research 55, no. 23, Suppl., 1995, 57715. 

Bocchi, V.; Foina, D.; Pochini, A.; Ungaro, R.; Andreetti, G. D., "Synthesis, ¹H NMR, ¹³C NMR Spectra and Conformational Preference of Open Chain Ligands on Lipophilic Macrocycles",  Tetrahedron, 1982, 38, 373.

Bollinger, J. E.; Mague, J. T.; Banks, W. A.; Kastin, A. J.; Roundhill, D. M., "Lipophilic Hexadentate Aluminum Complexes of New Phenolate-Derivatized Cyclohexanetriamine Ligands and Their Effect on Peptide Transport System (PTS-1)", Inorg. Chem., 1995, 34, 2143.

Bollinger, J. E.; Mague, J. T.; O^,Connor, C. J.; Banks, W. A.; Roundhill, D. M., "Lipophilic Hexadentate Gallium, Indium and Iron Complexes of New Phenolate-Derivatized Cyclohexanetriamine Ligands as Potential in vivo Metal Transfer Reagents", JCS, Dalton Trans., 1995, 1677.

Boston, D. R.; Rose, N. J., "An Encapsulation Reaction. Synthesis of the Clathro Chelate 1,8-Bis(fluoroboro)-2,7,9,14,15,20-hexaoxa-3,6,10,13,16,19-hexaaza-4,5,11,12,17,18-hexamethyl-bicyclo[6.6.6]eicosa-3,5,10,12,16,18-hexaenecobalt(III) Ion", J. Am. Chem. Soc., 1968, 90, 6859.

Bryant, L. H.; Yordanov, A.T.; Linnoila, J. J.; Brechbiel, M.W.; Frank, J. A., “First Noncovalently Bound Calix[4]arene- Gd^III - Albumin Complex”, Angew. Chem., Int. Ed. Engl., 2000, 39, 1641.

Busch, D. H., "The Compleat Coordination Chemistry-One Practitioner^,s Perspective", Chem. Revs. 1993, 93, 847.

Carty, A. J.; Tuck,  D. G., "The coordination chemistry of indium", Progr. Inorg. Chem., 1975, 19, 245.

Casnati, A.; Minari, P.; Pochini, A.; Ungaro, R., "Conformational Freezing of p-tert-Butylcalix[6]arene in the Cone Structure by Selective Functionalization at the Lower Rim: Synthesis of New Preorganized Ligands", JCS, Chem. Comm., 1991, 1413.

Chirakul, P.; Hampton, P.D.; Bencze, Z., “A Convergent Synthesis of Hexahomotriazacalix[3]arene Macrocycles”, J. Org, Chem., 2000, 65, 8297.

Cole, W. C.; DeNardo, S. J.; Meares, C. F. et. al., "Serum stability of copper-67 chelates: comparison with indium-111 and cobalt-57", Nucl. Med. Biol., 1986, 13, 363.

Collins, E. M.; Harris, S.; McKervey, M. A., "Molecular Receptors with the Calix[4]arene Substructure.  Synthesis of Derivatives with Mixed Ligating Functional Groups", JCS, Perkin Trans I, 1989, 372.

Cram, D. J.; Tanner, M. E.; Knobler, C. B., "Guest Release and Capture by Hemicarcerands Introduces the Phenomenon of Constrictive Binding", J. Am. Chem. Soc. 1991, 113, 7717.

Cram, D. J.; Blanda, M. T.; Paek, K.; Knobler, C. B., "Constrictive and Intrinsic Binding in a Hemicarcerand Containing Four Portals", J. Am. Chem. Soc., 1992, 114, 7765.

Cram, D. J.; Jaeger, R.; Deshayes, K., "Hemicarcerands That Encapsulate Hydrocarbons with Molecular Weights Greater than Two Hundred",  J. Am. Chem. Soc., 1993, 115, 10111.                  

Di Chiro, G., "New radiographic and isotopic procedures in neurological diagnosis", J. Am. Med. Assoc.,    1964, 188, 524.

Eisenwiener, K.-P.; Prata; M. I. M.; Buschmann, I.; Zhang, H.-W.; Santos, A. C.; Wenger, S.; Reubi, J. C.; Maecke, H. R., “NODAGATOC, a New Chelator-Coupled Somatostatin Analogue Labeled with [^67/68Ga ] and  [¹¹¹In ] for SPECT, PET, and Targeted Therapeutic Applications

            of Somatostatin Receptor(hsst2) Expressing Tumors”, Bioconjugate Chemistry 13, 2002, 530.

Ell, P. J.; Jarritt, P. H.; Costa, D. C. et al., "Functional imaging of the brain", Semin. Nucl. Med., 1987,    17, 214.

Green, M. A.; Welch, M. J.; Huffmann, J. C., "Synthesis and crystallographic characterization of a gallium salicylaldimine complex of radiopharmaceutical interest", J. Am. Chem. Soc., 1984, 106, 3689.

Gutsche, C. D.; Lin, L. -G., "The Synthesis of Functionalized Calixarenes", Tetrahedron, 1986, 42, 1633.

Gutsche, C. D. Calixarenes, Royal Society of Chemistry, Cambridge, U.K., 1989.

Gutsche, C. D.; Iqubal, M., "p-tert-Butylcalix[4]arene", Org. Synth., 1989, 68, 234.

Gutsche, C. D.; Dhawan, B.; Leonis, M; Stewart, D., "p-tert-Butylcalix[6]arene", Org. Synth., 1989, 68, 238.

Hampton, P.D.; Tang, W.; Wu,S.; Duesler, E.N., “Synthesis, X-ray structure and alkali-metal binding properties of a new hexahomotriazacalix[3]arene”, JCS; Perkin Trans.2, 1996, 1127.

Harris, W. R.; Pecoraro, V. L., "Thermodynamic binding constants for gallium transferrin", Biochemistry, 1988, 22, 292.

Hilfiker, K. A.; Brechliel, M. W.; Rogers, R. D.; Planalp, R. P., “Tricationic Metal Complexes ( [ML][NO₃]₃  M= Ga, In  ) of N, N’, N”-Tris(2-pyridyl-methyl)- cis- 1, 3, 5-triaminocyclohexane: Preparation and Structure ”, Inorg. Chem., 1997, 36, 4600.

Hillariet de Boisferon, M.; Raguin, O.; Thiercelin, C.; Dussaillant, M.; Rostene, W.; Barbet, J.; Pelegrin, A.; Gruaz-Guyon, A., “A.Improved Tumor Selectivity of Radiolabeled Peptides by Receptor and Antigen Dual Targeting in the Neurotensin Receptor Model”, Bioconjugate Chemistry 13, 2002,  654.

Hnatowich, D. J.; Layne, W. W.; Childs, R. L., "The preparation and labeling of DTPA-coupled albumin", Int. J. Appl. Radiat. Isot., 1982, 33, 327.

Hnatowich, D. J.; Virzi, F.; Doherty, P. W., "DTPA-coupled antibodies labeled with yttrium-90", J. Nucl. Med., 1985, 26, 503.           

Jurisson, S.; Berning, D.; Jia, W.; Ma, D., "Coordination Compounds in Nuclear Medicine", Chem. Rev., 1993, 93, 1137.

Kammerer, H.; Happel, G.; Bohmer, V.; Rathay, D., "Die stufenweise Synthese von 4, 11, 18, 25-Tetra-tert-butyl-[1.1.1.1]-metacyclophan-7, 14, 21, 28-tetraol und 4, 11-Dimethyl-[1.1.1.1]-metacyclophan-7, 14, 21, 28-tetraol",  Monatsch., 1978, 109, 767.

Koizumi, M.; Endo, K.; Kunimatsu, M. et. al., "Preparation of gallium-67-labeled antibodies using deferoxamine as a bifunctional chelate.  An improved method", J. Immunol. Methods, 1987, 104, 93.

Kuhl, D. E.; Edwards, R. Q., " Image separation radioisotope scanning", Radiology, 1963, 80, 653.

Kurihara, A; Deguchi, Y.; Pardridge, W. M., “Epidermal Growth Factor Radiopharmaceuticals: ¹¹¹In Chelation, Conjugation to a Blood-Brain Barrier Delivery Vector via a Biotin-Polyethylene Linker, Pharacokinetics, and in Vivo Imaging of Experimantal Brain Tumors”, Bioconjugate Chemistry, 1999, 10, 502.

Liu, S.; Rettig, S. J.; Orvig, C., "Polydentate Ligand Chemistry of Group 13 Metals: Effects of the Size and Donor Selectivity of Metal Ions on the Properties of Aluminum, Gallium and Indium Complexes with Potentially Heptadentate (N₄O₃) Amine Phenol Ligands",  Inorg. Chem., 1992, 31, 5400.

Loc'h, C.;  Maziere, B.; Comar, D., " A new generator for ionic gallium-68", J. Nucl. Med., 1980, 21, 171.

Martell, A. E.; Motekaitis, R. J., "Determination and Use of Stability Constants", VCH, New York, 1988.

Mathias, C. J.; Sun, Y.; Welch, M. J.; Green, M. A.; Thomas, J. A.; Wade, K. R.; Martell, A. E.,  “Targeting Radiopharmaceuticals: comparative biodistribution studies of gallium and indium complexes of multidentate ligands",  Nucl. Med. Biol., 1988, 15, 69.

Mathias, C. J.; Sun, Y.; Connett, J. M. et. al., "A new bifunctional chelate, BrMe2HBED:  an effective conjugate for radiometals and antibodies", Inorg. Chem., 1990, 29, 1475.

Moerlein, S. M.; Welch, M. J.; Raymond, K. N. et. al., "Tricatecholamide analogs of enterobactin as gallium- and indium-binding radiopharmaceuticals", J. Nucl. Med., 1981, 22, 710.

Moore, D. A.; Fanwick, P. E.; Welch, M. J., "Synthesis, characterization, and solid-state structure of a new hexachelating ligand and its complex with gallium (III)", Inorg. Chem., 1989, 28, 1504.

Moran, J. K.; Roundhill, D. M., "Introduction of Diphenylphosphinite Functional Groups onto Selected Positions on the Lower Rim of Calix[4]arenes and Calix[6]arenes", Inorg. Chem., 1992, 31, 4213.

Moran, J. K.; Georgiev, E. M.; Yordanov, A. T.; Mague, J. T.; Roundhill, D. M., "Synthesis and Structural Characterization of Calix[4]arenes, Calix[6]arenes and Calix[8]arenes with 3-Hydroxypropoxy or 2-Hydroxyethoxy Functional Groups Appended onto the Lower Rim", J. Org. Chem., 1994, 59, 5990.

Motekaitis, R. J.; Martell, A. E.; Welch, M. J., "Stabilities of trivalent metal complexes of phenolic ligands related to N,N'-bis(2-hydroxybenzyl)ethylenediamine-N,N'-diacetic acid (HBED)",  Inorg. Chem., 1990, 29, 1463.

Munch, J. H.; Gutsche, C. D., "p-tert-Butylcalix[8]arene", Org. Synth., 1989, 68, 243.

Noble, E. P.; Wurtman, A. J.; Axelrod, J., Life Sci. 1967, 6, 281.

Orvig, C.; Abrams, M. J., “Medical Inorganic Chemistry: Introduction”, Chem. Rev. (Washington, D.C.)    1999, 99, 2201.

Park, G.; Lu, F. H.; Ye, N.; Brechbiel, M. W.; Torti, S.V; Torti, F.M.; Planalp, R. P., “Novel iron complexes and chelators based on cis, cis- 1, 3, 5-triaminocyclohexane: iron- mediated ligand oxidation and biochemical properties”, J. Biol. Inorg. Chem., 1998, 3, 449.

Park, G.; Ye, N.; Rogers, R.D.; Brechbiel, M. W.; Planalp, R. P., “Effect of metal size on coordination geometry of N, N’, N” - tris (pyridylmethyl)- cis, cis- 1, 3, 5- triaminocyclohexane. Synthesis and structure of [M^IIL](ClO₄)₂    (M=Zn, Cd, and Hg)”, Polyhedron, 2000, 19, 1155.

Raichle, M. E., "Positron emission tomography tracer techniques", Adv. Chem. Ser., 1981, 197, 419.

Riccabona, G., "Neuro-nuclear medicine" 1990: what are the prospects?", Eur. J. Nucl. Med., 1991, 18, 1.

Roundhill, D. M.; Georgiev, E.; Yordanov, A., "Calixarenes with Nitrogen or Phosphorus Substituents on the Lower Rim", J. Incl. Phenom., 1994, 19, 101.

Roundhill, D. M., "Some new complexes of indium(III) with olefinic phosphines", J. Inorg. Nucl. Chem.,   1971, 33, 3367.

Roundhill, D. M., "Metal Complexes of Calixarenes", Prog. in Inorg. Chem., 1995, 43, 533.

Saha, G. B.; MacIntyre, W. J.; Go, R. T., "Cyclotrons and Positron Emission Tomography Radiopharmaceuticals for Clinical Imaging", Semin. Nucl. Med., 1992, 22, 150.

Schuhmacher, J.; Kaul, S.; Klivenyi, G.; Junkermann, H.; Magener, A.; Henze, M.; Doll, J.; Haberkorn, U.; Amelung, F.; Bastert, G., “Immunoscintigraphy with positron emission tomography: gallium-68 chelate imaging of breast cancer pretargeted with bispecific anti-MUC1/anti-Ga chelate antibodies”, Cancer research, 2001, 61, 3712.     

Setyawati, I. A.; Retting, S. J.; Orvig, C., “Cationic iron(III) complexes inth a hexadertate N₂ , N₂O₂ - aminopyridylphenolate ligand”, Canad. J. Chem., 1997, 77, 2033.        

Sheka ,I. A; Chaus, I.  S.; Mityureva, T. T., "The Chemistry of Gallium", Elsevier, New York, 1966.

Sillen, L. G.; Martell, A. E., "Stability Constants", Special Publication No. 17, The Chemical Society, London, 1964.

Sokoloff, L., "Relationships among local functional activity, energy metabolism, and blood flow in the central nervous system", Fed. Proc., 1981, 40, 2311.

Takemura, H. ; Shinmyozu, T.; Miura H.,; Khan, I.U.; Inazu, T., “Synthesis and Properties of N-Substituted Azacalix[n]arenes”, J. Incl. Phenom. Molec. Recogn. In Chem. 1994, 19, 193.

Taylor, M. J., "Comprehensive Coordination Chemistry", G. Wilkinson, ed., Pergamon, 1987, Vol. 3, Chapter 25.1.

Torti, S. V.; Torti, F. M.; Whitman, S. P.; Brechbiel, M.W.; Park, G.; Planalp, R. P., “Tumor Cell Cytotoxicity of a Novel Metal Chelator”, Blood, 1998, 92, 1384.

Tuck, D. G., "Critical Survey of Stability Constants of Complexes of Indium", Pure Appl. Chem., 1983, 55, 1477.

Vicens, J.; Böhmer, V., "Calixarenes:  A Versatile Class of Macrocyclic Compounds", Kluwer, Dordrecht, Netherlands, 1990.

Walovitch, R. C.: Hill,  T. C.; Garrity, S. T. et. al., "Characterization of technetium-99m-L,L-ECD for brain perfusion imaging, Part 1: pharmacology of technetium-99m ECD in nonhuman primates", J. Nucl. Med., 1988, 29, 5.

Wang, S.; Lee, R. J.; Mathias, C. J.; Green, M. A.; Low, P. S.,  “Synthesis, Purification, and Tumor Cell Uptake of  ⁶⁷Ga-Deferoxamine-Folate, a Potential Radiopharmaceutical for Tumor Imaging”, Bioconjugate Chem., 1996, 7, 56.     

Wang, S.; Luo, J.; Lantrip, D. A., Waters, D. J.; Mathias, C. J.; Green, M. A.; Fuchs, P. L;. Low, P. S. “Design and Synthesis of [¹¹¹In ]DTPA-Folate for Use as a Tumor-Targeted Radiopharmaceutical”, Bioconjugate Chemistry, 1997, 8, 673.

Washburn, L. C.; Sun, T. T. H.; Crook, J. E. et. al., "Yttrium-90-labeled monoclonal antibodies for cancer therapy",  Nucl. Med. Biol., 1986, 13, 453.

Wenzel, T. J.; Williams, E. J.; Haltiwanger, R. C.; et. al., “Studies of Metal Chelates with the Novel Ligand 2, 2, 7- Trimethyl- 3, 5- Octanedione”, Polyhedron, 1985, 4, 369.

67.        Variable Pressure and Temperature ¹H NMR Studies on Exchange Mechanisms in π-Allylpalladium(II) Complexes, R. L. Batstone-Cunningham, H. W. Dodgen, J. P. Hunt and D. M. Roundhill, J. Organometal. Chem., 289, 431 (1985).

72.        Axial Ligand Anation and Aquation Reactions in Diplatinum(III) Complexes.  Comparison of Aquation Rates between PtCl₆^2- and Diplatinum(III) Chloro Complexes having μ-Phosphato or μ-Pyrophosphito Ligands, R. El-Mehdawi, S. A. Bryan and D. M. Roundhill, J. Am. Chem. Soc., 107, 6282 (1985).

74.        Tetrakis(μ-pyrophosphito)diplatinum(II) tetraanion.  A Potential Inorganic Material for the Fabrication of Luminescent Solar Concentrators, D. M. Roundhill, Solar Energy, 36, 297 (1986).

83.        Synthesis and X-ray Structure of Na₁₀[Pt₂(μ-PO₄)₄(C₅H₃N₅O)₂]●22H₂O, a Complex with Doubly Deprotonated Guanine Anions Coordinated to Diplatinum(III), R. El-Mehdawi, F. R. Fronczek and D. M. Roundhill, Inorg. Chem., 25, 3714 (1986).

91.        Invited Paper (Stephenson Special Issue):  Diplatinum(II) and (III) Complexes with Bridging μ-(P,P')-Pyrophosphito or μ-(P,P')-Methylene- bisphosphinito Ligands.  Synthesis, Structure and NMR Analysis of the "Lantern" Complexes.  Crystal Structures of K₄[Pt₂(μ-CH₄O₄P₂)₄]●6H₂O and K₄[Pt₂(μ-CH₄O₄P₂)₄Cl₂]●8H₂O, C. King, D. M. Roundhill, M. K. Dickson and F. R. Fronczek. J. Chem. Soc., Dalton Trans., 2769 (1987).

111.      Chloride ion Substitution in η⁵-cpRuClL₂ Complexes, and the facile Cyclometalation of a Complexed Triphenyl phosphite at a Cationic Cyclopentadienyl Ruthenium Center, F. Joslin, J. T. Mague and D. M. Roundhill, Organometallics, 10, 521 (1991).

127.      Structure of Bis [(μ-chloro) bis (triphenylphosphine) palladium(II)] Bis [Tetrafluoroborate] Acetone. Dihydrate, S. Ganguly, E. M. Georgiev, J. T. Mague and D. M. Roundhill, Acta Crystallogr., C49, 1169 (1993).