Substituted porphyrins

Abstract

The present invention relates, in general, to a method of modulating physiological and pathological processes and, in particular, to a method of modulating cellular levels of oxidants and thereby processes in which such oxidants are a participant. The invention also relates to compounds and compositions suitable for use in such methods.

Description

TECHNICAL FIELD

The present invention relates, in general, to a method of modulating physiological and pathological processes and, in particular, to a method of modulating cellular levels of oxidants and thereby processes in which such oxidants are a participant. The invention also relates to compounds and compositions suitable for use in such methods.

BACKGROUND

Oxidants are produced as part of the normal metabolism of all cells but also are an important component of the pathogenesis of many disease processes. Reactive oxygen species, for example, are critical elements of the pathogenesis of diseases of the lung, the central nervous system and skeletal muscle. Oxygen free radicals also play a role in modulating the effects of nitric oxide (NO.). In this context, they contribute to the pathogenesis of vascular disorders, inflammatory diseases and the aging process.

A critical balance of defensive enzymes against oxidants is required to maintain normal cell and organ function. Superoxide dismutases (SODs) are a family of metalloenzymes that catalyze the intra- and extracellular conversion of O₂⁻ into H₂O₂ plus O₂, and represent the first line of defense against the detrimental effects of superoxide radicals. Mammals produce three distinct SODs. One is a dimeric copper- and zinc-containing enzyme (CuZn SOD) found in the cytosol of all cells. A second is a tetrameric manganese-containing SOD (Mn SOD) found within mitochondria, and the third is a tetrameric, glycosylated, copper- and zinc-containing enzyme (EC-SOD) found in the extracellular fluids and bound to the extracellular matrix. Several other important antioxidant enzymes are known to exist within cells, including catalase and glutathione peroxidase. While extracellular fluids and the extracellular matrix contain only small amounts of these enzymes, other extracellular antioxidants are also known to be present, including radical scavengers and inhibitors of lipid peroxidation, such as ascorbic acid, uric acid, and α-tocopherol (Halliwell et al, Arch. Biochem. Biophys. 280:1 (1990)).

The present invention relates generally to low molecular weight porphyrin compounds suitable for use in modulating intra- and extracellular processes in which superoxide radicals, or other oxidants such as hydrogen peroxide or peroxynitrite, are a participant. The compounds and methods of the invention find application in various physiologic and pathologic processes in which oxidative stress plays a role.

SUMMARY OF THE INVENTION

The present invention relates to a method of modulating intra- or extracellular levels of oxidants such as superoxide radicals, hydrogen peroxide, peroxynitrite, lipid peroxides, hydroxyl radicals and thiyl radicals. More particularly, the invention relates to a method of modulating normal or pathological processes involving superoxide radicals, hydrogen peroxide, nitric oxide or peroxynitrite, using low molecular weight antioxidants, and to methine (ie, meso) substituted porphyrins suitable for use in such a method. The substituted porphyrins are also expected to have activity as antibacterial and antiviral agents, and as ionophores and chemotherapeutics. Objects and advantages of the present invention will be clear from the description that follows.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1. Mechanism.

FIG. 2. Manganese meso-tetrakis-N-alkyl-pyridinium based porphyrins.

FIG. 3. SOD activity in vivo (E. coli) of 1, 2, 3* and 4* (20 μM) in minimal medium (mixture of atropoisomers, JI=SOD deficient strain, AB=parental strain).

FIG. 4. Structures of MnCl_xTE-2-PyP⁵⁺ (x=1 to 4).

FIG. 5. ¹H-NMR spectrum (porphyrin ring) of H₂Cl_2aT-2-PyP in CDCl₃ (δ=7.24 ppm). The four protons in alpha position of the four pyridyl nitrogens are taken as integration reference.

FIG. 6. Plot of the free energy of activation (ΔG^d) for the O₂′ dismutation reaction catalyzed by MnCl_xTE-2-PyP⁵+as a function of the ground state free energy change (ΔG^d) for MnCl_xTE-2-PyP⁵⁻ redox. ΔG^d and ΔG^o were calculated from k_cat and E^o_1/2 values reported in Table 4 (F, R, h and k_B are Faraday, molar gas, Planck and Boltzmann constants, respectively).

Numbers 0-4 correspond to x in MnCl_xTE-2-PyP⁵⁻

Corresponding data for one active site of Cu, Zn-SOD (Ellerby et al, J. Am. Chem. Soc. 1118:6556 (1996)).

FIG. 7. Illustrated are the chemical structures of three classes of antioxidants. A) The meso-porphyrin class is depicted where: R₁ is either a benzoic acid (tetrakis-(4-benzoic acid) porphyrin (TBAP)) or a N-methyl group in the 2 or 4 position of the pyridyl (tetrakis-(N-methylpyridinium-2(4)-yl) porphyrin (TM-2-PyP, TM-4-PyP)); R₂ is either a hydrogen (H) or a bromide (Br, OBTM-4-PyP) and where the porphyrin is ligated with either a manganese (Mn), cobalt (Co), iron (Fe), or zinc (Zn) metal. B) The vitamin E analog class is represented by trolox. C) The flavanoid class is represented by rutin.

FIG. 8. The time course of iron/ascorbate mediated oxidation of rat brain homogenates. Rat brain homogenates were incubated for various times with 0.25 μM FeCl₂ and 1 μM ascorbate, and lipid peroxidation was measured as thiobarbituric acid reactive species (TBARS) spectrophotometrically at 535 nm (n=3).

FIG. 9. The comparison of trolox (▪, rutin (▴), bovine CuZnSOD (●), MnOBTM-4-PyP (▾) and MnTM-2-PyP (♦) in their ability to inhibit iron/ascorbate mediated oxidation of rat brain homogenates. Rat brain homogenates were incubated for 30 minutes with 0.25 μM FeCl₂ and 1 μM ascorbate, and lipid peroxidation was measured as thiobarbituric acid reactive species. The amount of TBARS formed in 30 minutes was expressed as 100% lipid peroxidation (n=3-6). Sigmoidal dose response curves were derived from fitting the data to a non-linear regression program.

FIG. 10. The comparison of manganic (▴), cobalt (●), iron (▾) and zinc (▪) analogs of TBAP in their ability to inhibit iron/ascorbate mediated oxidation of rat brain homogenates. Rat brain homogenates were incubated for 30 minutes with 0.25 μM FeCl₂ and 1 μM ascorbate, and lipid peroxidation was measured as thiobarbituric acid reactive species. The amount of TBARS formed in 30 minutes was expressed as 100% lipid peroxidation (n=3−6). Sigmoidal dose response curves were derived from fitting the data to a non-linear regression program.

FIG. 11. The comparison of manganic (solid) and zinc (open) analogs of TM-4-PyP (squares) and TM-2-PyP (triangles) in their ability to inhibit iron/ascorbate mediated oxidation of rat brain homogenates. Rat brain homogenates were incubated for 30 minutes with 0.25 μM FeCl₂ and 1 μM ascorbate, and lipid peroxidation was measured as thiobarbituric acid reactive species. The amount of TBARS formed in 30 minutes was expressed as 100% lipid peroxidation (n=3−6). Sigmoidal dose response curves were derived from fitting the data to a non-linear regression program.

DETAILED DESCRIPTION OF THE INVENTION

The present invention relates to methods of protecting against the deleterious effects of oxidants, particularly, superoxide radicals, hydrogen peroxide and peroxynitrite, and to methods of preventing and treating diseases and disorders that involve or result from oxidant stress. The invention also relates methods of modulating biological processes involving oxidants, including superoxide radicals, hydrogen peroxide, nitric-oxide and peroxynitrite. The invention-further relates to compounds and compositions, including low molecular weight antioxidants (eg mimetics of scavengers of reactive oxygen species, including mimetics of SODs, catalases and peroxidases) and formulations thereof, suitable for use in such methods.

Mimetics of scavengers of reactive oxygen species appropriate for use in the present methods include methine (ie meso) β-substituted porphines, or pharmaceutically acceptable salts thereof. The invention includes both metal-free and metal-bound porphines. In the case of metal-bound porphines, manganic derivatives of methine (meso) β-substituted porphines are preferred, however, metals other than manganese, such as iron (II or III), copper (I or II), cobalt (II or III), or nickel (I or II), can also be used. It will be appreciated that the metal selected can have various valence states, for example, manganese II, III or V can be used. Zinc (II) can also be used even though it does not undergo a valence change and therefore will not directly scavenge superoxide. The choice of the metal can affect selectivity of the oxygen species that is scavenged. Iron-bound porphines, for example, can be used to scavenge NO. while manganese-bound porphines cannot. These metal bound porphines scavenge peroxynitrite; iron, nickel and cobalt bound porphines tend to have the highest reactivity with peroxynitrite.

Preferred mimetics of the invention are of Formula I or II:

or pharmaceutically acceptable salt thereof, wherein R is C₁-C₈ alkyl, preferably, C₁-C₄ alkyl, more preferably, methyl, ethyl or isopropyl, most preferably methyl. This mimetic can also be present metal-free or bound to a metal other than Mn. All atropoisomers of the above are within the scope of the invention, present in isolated form or as a mixture of at least two. Atropoisomers wherein at least 3, preferably 4, of the R groups are above the porphyrin ring plane can be particularly advantageous.

One or more of the pyrrole rings of the porphyrin of Formula I or II can be β-substituted at any or all beta carbons, ie: 2, 3, 7, 8, 12, 13, 17 or 18. Such substituents, designated P, can be an electron withdrawing group, for example, each P can, independently, be a NO₂ group, a halogen (eg Cl, Br or F), a nitrile, a vinyl group, or a formyl group. For example, there can be 1, 2, 3, 4, 5, 6, 7 or 8 halogen (eg Br) substituents (when there are less than 8 halogen substituents, the remaining P's are advantageously hydrogen). Such substituents alter the redox potential of the porphyrin and thus enhance its ability to scavenge oxygen radicals. Each P can, independently, also be hydrogen. When P is formyl, it is preferred that there be not more than 2 (on non adjacent carbons), more preferably 1, the remaining P's being hydrogen. When P is NO₂, it is preferred that there be not more than 4 (on non adjacent carbons), more preferably 1 or 2, the remaining P's being hydrogen.

Mimetics suitable for use in the present methods can be selected by assaying for SOD, catalase and/or peroxidase activity and stability. Mimetics can also be screened for their ability to inhibit lipid peroxidation in tissue homogenates using iron and ascorbate to initiate the lipid peroxidation and measuring the formation of thiobarbituric acid reactive species (TBARS) (Ohkawa et al, Anal. Biochem. 95:351 (1979) and Yue et al, J. Pharmacol. Exp. Ther. 263:92 (1992)). The selective, reversible and SOD-sensitive inactivation of aconitase by known O⁻₂ generators can be used as a marker of intracellular O⁻₂ generation. Thus, suitable mimetics can be selected by assaying for the ability to protect aconitase activity.

SOD activity can be monitored in the presence and absence of EDTA using the method of McCord and Fridovich (J. Biol. Chem. 244:6049 (1969)). The efficacy of a mimetic can also be determined by measuring the effect of the mimetic on the aerobic growth of a SOD null E. coli strain versus a parental strain lacking the specific mutations. Specifically, parental E. coli (AB1157) and SOD null E. coli. (JI132) can be grown in M9 medium containing 0.2% casamino acids and 0.2% glucose at pH 7.0 and 37° C.; growth can be monitored in terms of turbidity followed at 700 nm. This assay can be made more selective for SOD mimetics by omitting the branched chain, aromatic and sulphur containing amino acids from the medium (glucose minimal medium (M9), plus 5 essential amino acids) (see Example V).

Efficacy of active mimetics can also be assessed by determining their ability to protect mammalian cells against methylviologen (paraquat)-induced toxicity.

Specifically, rat L2 cells grown as described below and seeded into 24 well dishes can be pre-incubated with various concentrations of the SOD mimetic and then incubated with a concentration of methylviologen previously shown to produce an LC₇₅ in control L2 cells.

Efficacy of the mimetic can be correlated with a decrease in the methylviologen-induced LDH release (St. Clair et al, FEBS Lett. 293:199 (1991)).

The efficacy of SOD mimetics can be tested in vivo with mouse and/or rat models using both aerosol administration and parenteral injection. For example, male Balb/c mice can be randomized into 4 groups of 8 mice each to form a standard 2×2 contingency statistical model. Animals can be treated with either paraquat (40 mg/kg, ip) or saline and treated with SOD mimetic or vehicle control. Lung injury can be assessed 48 hours after paraquat treatment by analysis of bronchoalveolar lavage fluid (BALF) damage parameters (LDH, protein and % PMN) as previously described (Hampson et al, Tox. Appl. Pharm. 98:206 (1989); Day et al, J. Pharm. Methods 24:1 (1990)). Lungs from 2 mice of each group can be instillation-fixed with 4% paraformaldehyde and processed for histopathology at the light microscopic level.

Catalase activity can be monitored by measuring absorbance at 240 nm in the presence of hydrogen peroxide (see Beers and Sizer, J. Biol. Chem. 195:133 (1952)) or by measuring oxygen evolution with a Clark oxygen electrode (Del Rio et al, Anal. Biochem. 80:409 (1977)) Peroxidase activity can be measured spectrophotometrically as previously described by Putter and Becker: Peroxidases. In: Methods of Enzymatic Analysis, H. U. Bergmeyer (ed.), Verlag Chemie, Weinheim, pp. 286-292 (1983). Aconitase activity can be measured as described by Gardner and Fridovich (J. Biol. Chem. 266:19328 (1991)). The ability of mimetics to inhibit lipid peroxidation is assessed as described by Ohkawa et al (Anal. Biochem. 95:351 (1979)) and Yue et al (J. Pharmacol. Exp. Ther. 263:92 (1992)).

Active mimetics can be tested for toxicity in mammalian cell culture by measuring lactate dehydrogenase (LDH) release. Specifically, rat L2 cells (a lung Type II like cell; (Kaighn and Douglas, J. Cell Biol. 59:160a (1973)) can be grown in Ham's F-12 medium with 10% fetal calf serum supplement at pH 7.4 and 37° C.; cells can be seeded at equal densities in 24 well culture dishes and grown to approximately 90% confluence; SOD mimetics can be added to the cells at log doses (eg micromolar doses in minimal essential medium (MEM)) and incubated for 24 hours. Toxicity can be assessed by morphology and by measuring the release of the cytosolic injury marker, LDH (eg on a thermokinetic plate reader), as described by Vassault (In: Methods of Enzymatic Analysis, Bergmeyer (ed) pp. 118-26 (1983); oxidation of NADH is measured at 340 nm).

Synthesis of mimetics suitable for use in the present method can be effected using art-recognized protocols (see also Examples I, II, III and IV and Sastry et al, Anal. Chem. 41:857 (1969), Pasternack et al, Biochem. 22:2406 (1983); Richards et al, Inorg. Chem. 35:1940 (1996) and U.S. application Ser. No. 08/663,028, particularly the details therein relating to syntheses) Separation of atropoisomers can be effected using a variety of techniques.

One specific embodiment of the present invention relates to a method of regulating NO. levels by targeting the above-described porphines to strategic locations. NO. is an intercellular signal and, as such, NO. must traverse the extracellular matrix to exert its effects. NO., however, is highly sensitive to inactivation mediated by O⁻₂ present in the extracellular spaces. The methine (meso) β-substituted porphyrins of the invention can increase bioavalability of NO. by preventing its degradation by O⁻₂⁻.

In a further embodiment, the mimetics of the invention are used as catalytic scavengers of reactive oxygen species to protect against ischemia reperfusion injuries associated with myocardial infarction, stroke, acute head trauma, organ reperfusion following transplantation, bowel ischemia, hemorrhagic shock, pulmonary infarction, surgical occlusion of blood flow, and soft tissue injury. The mimetics can further be used to protect against skeletal muscle reperfusion injuries. The mimetics can also be used to protect against damage to the eye due to sunlight (and to the skin) as well as glaucoma, and macular degeneration in the eye. The mimetics can also be used to protect against and/or treat cataracts. The mimetics can also be used to protect against and/or treat inflammatory diseases of the skin (e.g., psoriasis). Diseases of the bone are also amenable to treatment with the mimetics. Further, connective tissue disorders associated with defects in collagen synthesis or degradation can be expected to be susceptible to treatment with the present mimetics, as should the generalized deficits of aging.

In yet another embodiment, the mimetics of the invention can be used as catalytic scavengers of reactive oxygen species to increase the very limited storage viability of transplanted hearts, kidneys, skin and other organs and tissues. The invention also provides methods of inhibiting damage due to autoxidation of substances resulting in the formation of O⁻₂ including food products, pharmaceuticals, stored blood, etc. To effect this end, the mimetics of the invention are added to food products, pharmaceuticals, stored blood and the like, in an amount sufficient to inhibit or prevent oxidation damage and thereby to inhibit or prevent the degradation associated with the autoxidation reactions. (For other uses of the mimetics of the invention, see U.S. Pat. No. 5,227,405). The amount of mimetic to be used in a particular treatment or to be associated with a particular substance can be determined by one skilled in the art.

In yet another embodiment, the mimetics of the invention can be used to scavenge hydrogen peroxide and thus protect against formation of the highly reactive hydroxyl radical by interfering with Fenton chemistry (Aruoma and Halliwell, Biochem. J. 241:273 (1987); Mello Filho et al, Biochem. J. 218:273 (1984); Rush and Bielski, J. Phys. Chem. 89:5062 (1985)). The mimetics of the invention may also be used to scavenge peroxynitrite, as demonstrated indirectly by inhibition of the oxidation of dihydrorhodamine 123 to rhodamine 123 and directly by accelerating peroxynitrite degradation by stop flow analysis.

Further examples of specific diseases/disorders appropriate for treatment using the mimetics of the present invention include diseases of the central nervous system (including AIDS dementia, stroke, amyotrophic lateral sclerosis (ALS), Parkinson's disease and Huntington's disease) and diseases of the musculature (including diaphramic diseases (eg respiratory fatigue in emphysema, bronchitis and cystic fibrosis), cardiac fatigue of congestive heart failure, muscle weakness syndromes associated with myopathies, ALS and multiple sclerosis). Many neurologic disorders (including stroke, Huntington's disease, Parkinson's disease, ALS, Alzheimer's and AIDS dementia) are associated with an over stimulation of the major subtype of glutamate receptor, the NMDA (or N-methyl-D-aspartate) subtype. On stimulation of the NMDA receptor, excessive neuronal calcium concentrations contribute to a series of membrane and cytoplasmic events leading to production of oxygen free radicals and nitric oxide (NO.). Interactions between oxygen free radicals and NO. have been shown to contribute to neuronal cell death. Well-established neuronal cortical culture models of NMDA-toxicity have been developed and used as the basis for drug development. In these same systems, the mimetics of the present invention inhibit NMDA-induced injury. The formation of O⁻₂ radicals is an obligate step in the intracellular events culminating in excitotoxic death of cortical neurons and further demonstrate that the mimetics of the invention can be used to scavenge O⁻₂ radicals and thereby-serve as protectants against excitotoxic injury.

The present invention also relates to methods of treating AIDS. The NfKappa B promoter is used by the HIV virus for replication. This promoter is redox sensitive, therefore, an antioxidant can regulate this process. This has been previously shown for two metalloporphyrins distinct from those of the present invention (Song et al, Antiviral Chem. And Chemother. 8:85 (1997)). The invention also relates to methods of treating arthritis, systemic hypertension, atherosclerosis, edema, septic shock, pulmonary hypertension, including primary pulmonary hypertension, impotence, MED, infertility, endometriosis, premature uterine contractions, microbial infections, gout and in the treatment of Type I and Type II diabetes mellitus. The mimetics of the invention can be used to ameliorate the toxic effects associated with endotoxin, for example, by preserving vascular tone and preventing multi-organ system damage.

Inflammations, particularly inflammations of the lung, are amenable to treatment using the present invention (note particularly the inflammatory based disorders of asthma, ARDS including oxygen toxicity, pneumonia (especially AIDS-related pneumonia), cystic fibrosis, chronic sinusitis and autoimmune diseases (such as rheumatoid arthritis)). EC-SOD is localized in the interstitial spaces surrounding airways and vasculature smooth muscle cells. EC-SOD and O₂⁻ mediate the antiinflammatory—proinflammatory balance in the alveolar septum. NO. released by alveolar septal cells acts to suppress inflammation unless it reacts with O₂⁻ to form ONOO⁻. By scavenging O₂⁻, EC-SOD tips the balance in the alveolar septum against inflammation. Significant amounts of ONOO⁻ will form only when EC-SOD is deficient or when there is greatly increased O₂⁻release. Mimetics described herein can be used to protect against destruction caused by hyperoxia.

The invention further relates to methods of treating memory disorders. It is believed that nitric oxide is a neurotransmitter involved in long-term memory potentiation. Using an EC-SOD knocked-out mouse model (Carlsson et al, Proc. Natl. Acad. Sci. USA 92:6264 (1995)), it can be shown that learning impairment correlates with reduced superoxide scavenging in extracellular spaces of the brain. Reduced scavenging results in higher extracellular O⁻₂ levels. O⁻₂ is believed to react with nitric oxide thereby preventing or inhibiting nitric oxide-medicated neurotransmission and thus long-term memory potentiation. The mimetics of the invention can be used to treat dementias and memory/learning disorders.

The availability of the mimetics of the invention also makes possible studies of processes mediated by O₂⁻, hydrogen peroxide, nitric oxide and peroxynitrite.

The mimetics described above can be formulated into pharmaceutical compositions suitable for use in the present methods. Such compositions include the active agent (mimetic) together with a pharmaceutically acceptable carrier, excipient or diluent. The composition can be present in dosage unit form for example, tablets, capsules or suppositories. The composition can also be in the form of a sterile solution suitable for injection or nebulization. Compositions can also be in a form suitable for opthalmic use. The invention also includes compositions formulated for topical administration, such compositions taking the form, for example, of a lotion, cream, gel or ointment. The concentration of active agent to be included in the composition can be selected based on the nature of the agent, the dosage regimen and the result sought.

The dosage of the composition of the invention to be administered can be determined without undue experimentation and will be dependent upon various factors including the nature of the active agent, the route of administration, the patient, and the result sought to be achieved. A suitable dosage of mimetic to be administered, for example, IV or topically, can be expected to be in the range of about 0.01 to 100 mg/kg/day, preferably 0.1 to 10 mg/kg/day. For aerosol administration, it is expected that doses will be in the range of 0.01 to 1.0 mg/kg/day. Suitable doses of mimetics will vary, for example, with the mimetic and with the result sought. The results of Faulkner et al (J. Biol. Chem. 269:23471 (1994)) indicate that the in vivo oxidoreductase activity of the mimetics is such that a pharmaceutically effective dose will be low enough to avoid problems of toxicity. Doses that can be used include those in the range of 1 to 50 mg/kg.

Certain aspects of the present invention will be described in greater detail in the non-limiting Examples that follow.

EXAMPLES

The following chemicals were utilized in Examples I-V that follow.

The chloride salts of ortho and meta metal-free ligands (H₂TM-2-PyPCl₅ and H₂TM-3-PyPCl₅) were purchased from MidCentury Chemicals, and the tosylate salts of the para metal-free ligand H₂TM-4-PyP(CH₅PhSO₃)₅) were purchased from Porphyrin Products. The purity was checked in terms of elemental analysis and spectral properties, ie, molar absorptivities and corresponding wave-length of the Soret bands. The Soret band properties of metal-free ligands were ε_413.3nm=2.16×10⁵M⁻¹cm⁻¹ (H₂TM-2-PyPCl₄), ε_{416.6 nm}=3.18×10⁵ M⁻¹cm⁻¹ (H₂TM-3-PyPCl₄), ε_{422.0 nm}=2.35×10⁵ M⁻¹cm⁻¹ (H₂TM-4-PyPCl₄). The non-methylated ortho metal-free ligand (H₂T-2-PyP) was bought from MidCentury Chemicals and the purity checked in terms or elemental analysis (see below). Iodoethane, 1-iodobutane, anhydrous manganese chloride (MnCl₂), MnCl₂.4H₂O, tetrabutylammonium chloride (TBA) and ammonium hexafluorophosphate (PF₆NH₄) were purchased from Aldrich.

EXAMPLE I

Synthesis of meso-tetrakis-(N-methylpyridinium-2-yl)porphyrin and meso-tetrakis-(N-methylpyridinium-3-yl)porphyrin

Metal-free porphyrins meso-tetrakis-(2-pyridyl)porphyrin (H₂T-2-PyP) and meso-tetrakis-(3-pyridyl)porphyrin (H₂T-3-PyP) were synthesized via Rothmund condensation with use of a modified Adler procedure (Kalyanasundaram, Inorg. Chem. 23:2453 (1984); (Torrens et al, J. Am. Chem. Soc. 94:4160 (1972)). Into a 100 mL refluxing solution of propionic acid were slowly injected equimolar amounts of freshly distilled pyrrole and pyridine-2- or pyridine-3-carboxyladehyde, and the solution was allowed to reflux for about 45 min, after which the propionic acid was distilled off. The black residues were neutralized with NaOH, washed with methanol, dissolved in CH₂Cl₂ (dichloromethane) and chromotographed on a neutral Woelm alumina column prepared with acetone. After elution of a pale blue fraction, H₂TPyP was eluted with the use of CH₂Cl₂ containing 5-10% of pyridine. Shiny dark purple crystals were recovered from the dark red eluant after removal of solvents on rotavaporator. Methylation of H₂TPyPs was carried using the excess of methyl-p-toluensulfonate in refluxing chloroform (Kalyanasundaram, Inorg. Chem. 23:2453 (1984); (Hambright et al, Inorg. Chem. 15:2314 (1976′)). Both of the alkylated porphyrins spontaneously precipitated from hot chloroform solutions and were washed with ether and air dried.

EXAMPLE II

Preparation of Manganese Complexes of Ortho, Meta and Para Isomers of H₂TMPyP⁴⁺

The metallation was performed in water at room temperature. The porphyrin to metal ratio was 1:5 in the case of meta and ortho isomers and 1:14 in the case of para isomer. The solid MnCl₂×4H₂O (Aldrich) was added to the aqueous metal-free porphyrins after the pH of the solution was brought to ˜pH=10.2. The metallation was completed inside an hour in the cases of all three isomers. For the preparation of ortho and meta compounds, MnTM-2-PyP⁵⁺ and MnTM-3-PyP⁵⁺+, 300 mg of the metal-free ligand, either H₂TM-2-PyP 4 or H₂TM-3PyP⁴⁺, was dissolved in 100 mL water, pH brought to 10.2 with several drops of 1M NaOH, followed by the addition of 340 mg of MnCl₂. The metallation was followed spectrally through the disappearance of the Soret band of H₂TM-2-PyP⁴⁺ or H₂TM-3-Pyp⁴⁺ at 413.3 nm or 416.6 nm, respectively, and the appearance of the Soret bands of manganese complexes at 454.1 nm and 459.8 nm, respectively.

The excess of metal was eliminated as follows for all three (ortho, meta and para) isomers of MnTMPyP⁵⁺. The MnTMPyP⁵⁺ was precipitated as PF₆ salt by adding 50-fold excess of NH₄PF₆. The precipitate was washed with 2-propanol:diethylether=1:1, and dried in vacuum at room temperature. Dry PF₆⁻ salt of MnTMPyP⁵⁻ was then dissolved in acetone (370 mg in 100 mL acetone) and 1 g of tetrabutylammonium chloride added. The precipitate was washed with acetone and dried overnight in vacuum at room temperature. In order to obtain a pure compound, the procedure was repeated. The elemental analysis was done for all metallated isomers. The compounds were analyzed in spectral terms and the following data were obtained: Soret bands properties of metallated compounds were: ε_{4540.1 nm}=12.3×10⁴ cm⁻¹ M⁻¹ (MnTM-2-PyPCl₅), ε_{459.8 nM}=13.3×10⁴ cm⁻¹ M⁻¹ (MnTM-3-PyPCl₅), ε_{462.2 nm}=13.9×10⁴ cm⁻¹ M⁻¹ (MnTM-4-PyPCl₅).

Metallation was performed in methanol as well. In addition, when performed in water, the metal:ligand ratio varied from 1:5, to 1:14 to 1:100. Under all conditions, the given molar absorptivities were obtained. The calculations were based on the metal-free ligands that were analyzed prior to metallation. The molar absorptivities of the metal-free ligands were consistent with literature as well as their elemental analyses.

The elemental analyses of MnTM2-PyPCl₅ and MnTM-3-PyPCl₅ are shown in Table 1

	TABLE 1
	C*	H*	N*
MnTM2-PyPCl5.6 H2O	52.99(52.90)	4.85(4.64)	11.22(11.21)
MnTM-3-PyPCl5.3	55.41(54.87)	4.97(4.40)	11.10(11.69)
H2O
*Found (calcd).

EXAMPLE III

Synthesis of manganic meso-tetrakis-(N-ethylpyridinium-2-yl) porphyrin

50 mg of H₂T-2-PyP was dissolved in 30 mL of anhydrous dimethylformamide (DMF) and the solution was stirred and heated at 100° C. 20 mg of anhydrous MnCl₂ (20 eq) were added and the solution stirred for 3 days. The completion of the metallation was checked by UV spectroscopy. Upon metallation, the temperature was decreased to 60° C., 0.65 mL of iodoethane (100 eq) was added, and the solution was stirred for 7 days (Perree-Fauvet et al, Tetrahedron 52:13569 (1996)). DMF was evaporated, 10 mL of acetone was added, and the product was precipitated adding 20 mL of a solution of TEA in acetone (0.45 M); indeed, contrary to the iodide salt, the chloride salt precipitates in acetone. The product was purified using the “double precipitation” method, as described above. The product was dried overnight in vacuum, over P₂O₅, at 70° C., leading to 125 mg (95%) of a dark purple solid. UV (H₂O), ε_{454.0 nm}=1.41×10⁵ M⁻¹ cm⁻¹. Elemental analysis, calcd. for MnC₄₈N₈H₄₄Cl₅.5H₂O: C (54.64), H (5.16), N (10.62); found: C (54.55), H (5.36), N-(10.88).

EXAMPLE IV

Synthesis of manganic meso-tetrakis-(N-butylpyridinium-2-yl)porphyrin

The same procedure described above was used. 0.92 mL of 1-iodobutane (100 eq) was added and the mixture stirred at 100° C. for 7 days. Drying of the chloride salt resulted in 70 mg (50%) of a dark purple viscous product. The elemental analysis was thus performed on the hexafluorophosphate salt (non-viscous) The chlorine salt is water-soluble (micelles were not observed). UV(H₂O) of the chloride salt, ε_454.0 1.21×10⁵M⁻¹ cm⁻¹. Elemental analysis, calcd. For MnC₅₆H₆₀N₈P₅F₃₀.H₂O: C(40.94), H(3.80), N(6.82); found: C(41.15), H(4.35), N(6.52).

EXAMPLE V

The ortho effect makes manganic meso-tetrakis-(N-alkylpyridinium-2-yl)-porphyrin a powerful superoxide dismutase mimic

The superoxide dismutase activity of the mimetics of the invention depends on a number of factors, including thermodynamic factors (eg the metal-centered redox-potential see FIG. 1)), and kinetic factors (eg electrostatic facilitation). In an in vitro enzymatic assay of SOD activity (see McCord and Fridovich, J. Biol. Chem. 244:6049 (1969)), the ortho compound “3” proves to be more than an order of magnitude more active than the para compound “1” (see FIG. 2 (note also Table 2 where “2” is the meta compound and “4” and “5” are ortho compounds that carry 4 ethyl or 4 butyl groups, respectively)).

The activity in vivo of the mimetics of the invention can be tested on an E. coli strain deleted of the genes coding for both the MnSOD and FeSOD. In this assay, the efficacy of a mimetic is determined by measuring the effect of the mimetic on the aerobic growth of a SOD null E. coli strain versus a parental strain. Specifically, parental E. coli (AB1157) and SOD null E. coli. (JI132) are grown in M9 medium containing 0.2% casamino acids and 0.2% glucose at pH 7.0 and 37° C.; growth is monitored in terms of turbidity followed at 700 nm. This assay is made more selective for SOD mimetics by omitting the branched chain, aromatic and sulphur containing amino acids from the medium (glucose minimal medium (M9), plus 5 essential amino acids). As shown in FIG. 3, the increase in activity by the “ortho effect” was confirmed in that, under these growth conditions, SOD null cells cultured in the presence of compound “1” did not show an increase in A₇₀₀ while such cells cultured in the presence of compounds “3” and “4” (as well as “2”) did.

The “ortho effect” also decreases the toxicity. It is well known that porphyrins, and particularly cationic porphyrins, interact with DNA and can act as DNA cleavers. This fact can be an issue in the use of metallo-porphyrins as anti-tumor drugs. The present mimetics avoid this interaction. In addition to the increase in activity, the interaction with DNA of the meta “2” and the ortho “3” compounds, is greatly decreased. This is clearly demonstrated by the measurements of the SOD activity in vitro in the presence of DNA (see Table 2), and by the decreased toxicity in vivo (E. coli) (see FIG. 3).

In order to maximize the decrease in toxicity due to interaction with DNA, two derivatives of the ortho compound have been prepared which carry four ethyl or four butyl groups (“4” and “5”, respectively). The ethyl derivative “4” was significantly less toxic than the methyl derivative “3” (see Table 2 and FIG. 3). However, in comparison to the ethylated derivative “4”, the butylated derivative did not show a further decrease in toxicity (see Table 2). These data indicate that ortho ethyl groups are sufficient to inhibit binding of the porphyrin to DNA.

TABLE 2
Table. UV parameters, redox potential (vs NHE), SOD like
activity and DNA interaction parameters of 1, 2, 3 and
its atropisomers, 4 and 5 (*mixture of atropisomers, δSB =
Soret band wave-lenght, ε = molecular absortivity of the
Soret band, E1/2 = one-electron metal-centered redox-
potential. k = rate constant for the superoxide
dismutation reaction, DNA-IC50 = concentration of DNA
for 50% inhibition of the superoxide dismutation reaction).
	δSB (nm)	ε(103)	E1/2 (V)	k (M−1S−1)	DNA-IC50
1	462.2	139	+0.060	3.8 104	7.0 10−4
2	459.8	133	+0.042	4.1 104	2.2 10−5
3*	454.0	123		4.5 107	3.3 10−5
4*	454.0	141		4.5 107	6.7 10−5
5*	454.0	120		3.0 107	6.7 10−1

EXAMPLE VI

Syntheses and Superoxide Dismutating Activities of Partially (1 to 4) β-Chlorinated Derivatives of Manganese (III) Meso-tetrakis-(N-ethylpyridinium-2-yl)-Porphyrin

Materials and Methods

Materials. 5,19,15,20-Tetrakis-(2-pyridyl)-porphyrin (H₂T-2-PyP) was purchased from Mid-Century chemicals (Posen, Ill.) (Torrens et al, J. Am. Chem. Soc. 94:4160 (1972)). N-Chlorosuccinimide (NCS), ethyl-p-toluenesulfonate (ETS), tetrabutylammonium chloride (98%) (TBAC), ammonium hexafluorophosphate (NH₄PF₆), manganese chloride, sodium L-ascorbate (99%), cytochrome c, xanthine, ethylenedinitrilotetraacetic acid (EDTA), N,N-dimethylformamide (98.8%, anhydrous) and 2-propanol (99.5%) were from Sigma-Aldrich. Ethanol (absolute), acetone, ethyl ether (anhydrous), chloroform and dichloromethane (HPLC grade) were from Mallinckrodt, and used without further purification. Xanthine oxidase was supplied by R. D. Wiley (Waud et al, Arch. Biochem. Biophys. 19:695 (1975)). Thin-layer chromatography (TLC) plates (Baker-flex silica gel IB) were from J. T. Baker (Phillipsburg, N.J.). Wakogel C-300 was from Wako Pure Industry Chemicals, Inc (Richmond, Va.).

Instrumentation. Proton nuclear magnetic resonance (¹H-NMR) spectra were recorded on a Varian Inova 400 spectrometer. Ultravisible/visible (UV/VIS) spectra were recorded on a Shimadzu spectrophotometer Model UV-260. Matrix-assisted laser desorption/ionization—time of flight—(MALDI-TOFMS) and electrospray/ionization (ESMS) mass spectrometry were performed on a Bruker Proflex III™ and a Fisons Instruments VG Bio-Q triple quadrupole spectrometers, respectively.

H₂Cl₁T-2-PyP. 50 mg (8.1×10⁻⁵ moles) of H₂T-2-PyP was reflexed in chloroform with 43 mg (3.22×10⁻¹ moles) of NCS (Ochsenbein et al, Angew. Chem. Int. Ed. Engl. 33:348 (1994). The reaction was followed by normal phase silica TLC using a mixture EtOH/CH₂Cl₂ (5:95) as eluant. After 6 hours of reaction the solution was washed once with distilled water. The chloroform was evaporated and the products of the reaction were chromatographed over 100 g of Wakogel C-300 on a 2.5×50 cm column using the same eluant. The fraction corresponding to H₂Cl₁T-2-PyP was purified again using the same system leading to 16 mg of a black purple solid (30%). TLC: R^f=0.47. UV/VIS (CHCl₃): λ_nm (logε) 419.6 (5.44), 515.2 (4.21), 590.0 (3.72), 645.8 (3.25). MALDI-TOFMS: m/z=654 (M+H⁻). ¹H-NMR (CDCl₃): δ_ppm-2.91 (2H, NH); 7.66-7.74 (m, 4H); 7.99-8.21 (m, 8H); 7.68 (s, 1H); 8.74 (d, 1H, J 6 Hz); 8.76 (d, 1H, J 6 Hz); 8.76 (d, 1H, J 6 Hz); 8.88 (d, 1H, J 6 Hz); 8.90 (d, 1H, J 6 Hz); 8.94 (d, 1H, J 6 Hz); 9.04-9.14 (m, 4H).

H₂C_2aT-2-PyP. The same procedure as described above, leading to 5.3 mg of a black purple solid (10%). TLC: R_f=0.50. UV/VIS (CHCl₃): λ_nm (logε) 421.4 (5.38), 517.8 (4.21), 591.4 (3.78), 647.6 (3.51). MALDI-TOFMS: m/z=688 (M+H⁻). ¹H-NMR (CDCl₃): δ_ppm-2.98 (2H, NH); 7.66-7.74 (m, 4H); 8.00-8.20 (m, 8H); 8.70 (s, 2H); 8.82 (d, 2H, J 6 Hz); 8.91 (d, 2H, J 6 Hz); 9.06-9.14 (m, 4H).

H₂Cl_2b−2cT-2-PyP. The same procedure leading to 11 mg of a black purple solid (20%). TLC: R_f=0.53. UV/VIS (CHCl₃): A (logε) 421.4 (5.42), 516.8 (4.25), 593.2 (3.74), 646.2 (3.31); MALDI-TOFMS, m/z=688 (M+H⁻). ¹H-NMR (CDCl₃): δ_ppm-3.04 (2H, NH); -2.84 (1H, NH); -2.87 (1H, NH; 7.66-7.74 (m, 8H); 7.98-8.20 (m, 16H); 8.59 (s, 1H); 8.61 (s, 1H); 8.73 (d, 2H, J<2 Hz); 8:78 (d, 2H, J 6 Hz); 8.87 (d, 2H, J 6 Hz); 8.93 (d, 2H, J<2 Hz); 9.02-9.14 (m, 8H).

H₂Cl₃T-2-PyP. The same procedure using 65 mg (4.87×10⁻⁴ moles) of NCS, leading to 8.4 mg of a black purple solid (14%). TLC: R_f=0.55. UV/VIS (CHCl₃): λ_nm (logε) 422.8 (5.37), 519.4 (4.21), 593.8 (3.71), 651.4 (3.37). MALDI-TOFMS: m/z=723 (M+H⁻). ¹H-NMR (CDCl₃): δ_ppm-3.08 (1H, NH); -3.15 (1H, NH); 7.66-7.74 (m, 4H); 8.00-8.18 (m, 8H); 8.56 (s, 1H), 8.72 (d, 1H, J 6 Hz); 8.76 (d, 1H, J 6 Hz); 8.82 (d, 1H, J 6 Hz); 8.88 (d, 1H, J 6 Hz); 9.04-9.14 (m, 4H).

H₂Cl₄T-2-PyP. The same procedure using 65 mg (4.87×10⁻¹ moles) of NCS, leading to 7.3 mg of a black purple solid (12%). TLC: R_f=0.58. UV/VIS (CHCl₃): λ_nm (logε) 423.4 (5.33), 520.0 (4.19), 595.6 (3.66), 651.0 (3.33). MALDI-TOFMS: m/z=758 (M+H⁻). ¹H-NMR (CDCl₃): δ_ppm-3.14 (2H, NH); 7.66-7.74 (m, 4H); 7.98-8.16 (m, 8H); 8.74 (d, 4H, J<2 Hz); 9.06-9.12 (m, 4H).

MnTE-2-PyP⁵⁻. 100 mg (1.62×10⁻¹ moles) of H₂T-2-PyP was dissolved in 5 mL of warm DMF (anhydrous), 5.5 mL (3.22×10⁻² moles) of ethyl-p-toluenesulfonate (ETS) was added under stirring at 90° C. and allowed to react for 24-48 hours. The completion of tetra-N-ethylation was followed by normal phase silica TLC using a mixture KNO_3sat/H₂O/CH₃CN (1:1:8) as eluant (Batinic-Haberle et al, J. Biol. Chem. 273:24521 (1998)). Upon the completion of the reaction, the DMF was removed in vacuo and 5 mL of acetone was then added. To this solution, a concentrated solution of tetrabutylammonium chloride (TBAC) in acetone (˜1 g/10 mL acetone) was added dropwise under stirring until precipitation of the chloride was complete. The resulting purple solid was dissolved in 10 mL of water, the pH of the solution was raised to 12 with NaOH and 640 mg of MnCl₂4H₂O (3.23×10⁻³ moles) was added (Batinic-Haberle et al, J. Biol. Chem. 273:24521 (1998). Upon completion of metallation, the pH was lowered between 4 and 7 in order to facilitate the auto-oxidation of Mn (II) into Mn (III), and the excess of metal was eliminated as follows. The solution was filtered, and a concentrated aqueous solution of NH₄PF₆ was added to precipitate the metalloporphyrin as the PF₆⁻ salt (Batinic-Haberle et al, Arch. Biochem. Biophy. 343:225 (1997); Richards et al, Inorg. Chem. 35:1940 (1996)). The precipitate was thoroughly washed with a mixture 2-propanol/ethyl ether (1:1), dried in vacuo at room temperature. The resulting solid was then dissolved in acetone and a concentrated solution of TBAC was added to isolate the metalloporphyrin in the form of its chloride salt. The precipitate was washed thoroughly with acetone and dried in vacuo at room temperature leading to 150 mg of a black red solid (95%). TLC: R_f=0.18. UV/VIS (H₂O): λ_nm (logε) 364.0 (4.64), 453.8 (5.14), 558.6 (4.05). ESMS: m/z=157.4 (M⁵⁻/5). Anal. calcd. for MnC₄₈N₈H₄₄Cl₅H₂O: C. 54.64; H. 5.16; N. 10.62. Found: C. 54.55; H. 5.40; N. 10.39. (See FIG. 4 for compound structures).

MnCl₁TE-2-PyP⁵⁻. The same procedure as described above starting from 10 mg (1.53×10⁻⁵ moles) of H₂Cl₁T-2-PyP and 0.5 mL (2.94×10⁻³ moles) of ETS in 1 mL of DMF. TLC: R_f=0.20. UV/VIS (H₂O): λ_nm (logε) 365.6 (4.63), 455.6 (5.13), 560.6 (4.02). ESMS: m/z=164.3 (M⁵⁻/5). Anal. calcd. for MnC₄₈N₈H₄₃Cl₆5H₂O: C. 52.91; H. 4.90; N. 10.28. Found: C. 52.59; H. 5.28; N. 10.14.

MnCl_2aTE-2-PyP⁵⁻. The same procedure starting from 5 mg (7.28×10⁻⁶ moles) of H₂Cl_2aT-2-PyP and 0.25 mL (1.47×10⁻³ moles) of ETS, leading to 7.5 mg of a black red solid (95%). TLC: R_f=0.21. UV/VIS (H₂O): λ_nm (logε) 365.8 (4.58), 456.4 (5.05), 562.2 (4.00). ESMS: m/z=171.1 (M⁵⁺/5). Anal. calcd. for MnC₄₈N₈H₄₂Cl₇6H₂O: C. 50.48; H. 4.77; N. 9.81. Found: C. 50.08; H. 4.60; N. 10.01.

MnCl_2b−2cTE-2-PyP⁵⁺. The same procedure starting from 5 mg (7.28×10⁻⁶ moles) of H₂Cl_2b−2cT-2-PyP, leading to 7.5 mg of a black red solid (95%). TLC: R_f=0.22. UV/VIS (H₂O): λ_nm (logε) 365.2 (4.63), 457.4 (5.08), 462.2 (4.06). ESMS: m/z=171.1 (M⁵⁺/5). Anal. calcd. for MnC₄₈N₈H₄₂Cl₇5H₂O: C. 51.29; H. 4.66; N. 9.97. Found: C. 51.31; H. 5.19; N. 9.68.

MnCl₃TE-2-PyP⁵⁻. The same procedure starting from 5 mg (6.93×10⁻⁶ moles) of H₂Cl₃T-2-PyP, leading to 7.5 mg of a black brown solid (95%). TLC: R_f=0.23. UV/VIS (H₂O): λ_nm (logε) 364.8 (4.58), 458.0 (4.98), 466.4 (4.00). ESMS: m/z=178.1 (M⁵⁻/5). Anal. calcd. for MnC₄₈N₈H₄₁Cl₈6H₂O: C. 49.00; H. 4.54; N. 9.52. Found: C. 48.40; H. 4.26; N. 9.59.

MnCl₄TE-2-PyP⁵⁻. The same procedure starting from 5 mg (6.61×10⁻⁶ moles) of H₂Cl₄T-2-PyP, leading to 7.5 mg of a black brown solid (95%). TLC: R, =0.24. UV/VIS (H₂O): λ_nm (logε) 365.8 (4.52), 459.2 (4.90), 567.0 (3.96). ESMS: m/=184.9 (M⁵⁻/5). Anal. calcd. for MnC₄₈N₈H₄₀Cl₉5H₂O: C. 48.33; H. 4.22; N. 9.39. Found: C. 48.38; H. 4.45; N. 9.53.

Electrochemistry. The electrochemical characterization was performed as described previously on a Voltammetric Analyzer Model 600 (CH instrument) using a glassy carbon electrode (Ag/AgCl reference and Pt auxiliary electrodes), at 0.5 mM porphyrin, pH 7.8 (0.05 M phosphate buffer), 0.1 M NaCl. The potentials were standardized against potassium ferricyanide/potassium ferrocyanide couple (Batinic-Haberle et al, Arch. Biochem. Biophys. 343:225 (1997); Kolthof et al, J. Phys. Chem. 39:945 (1974)).

Superoxide dismuting activity. The SOD-like activities were measured using the xanthine/xanthine oxidase system as a source of O₂⁻ and ferricytochrome c as its indicating scavenger (McCord et al, J. Biol. Chem. 244:6049 (1969)). O₂⁻ was produced at the rate of 1.2 μM per minute and reduction of ferricytochrome c was followed at 550 nm. Assays were conducted in presence of 0.1 mM EDTA in 0.05 M phosphate buffer (pH 7.8). Rate constants for the reaction of the compounds were based upon competition with 10 μM cytochrome c, k_{cyr c}=2.6×10⁵ M⁻¹s⁻¹ (Butler et al, J. Biol. Chem. 257:10747 (1982)). All measurements were done at 25° C. Cytochrome c concentration was at least 10³-fold higher than the concentrations of the SOD mimics and the rates were linear for at least two minutes, during which the compounds intercepted ˜100 equivalents of O₂⁻, thus confirming the catalytic nature of O₂⁻ dismutation in presence of the mimics.

Results

Despite increasing knowledge on the purification of water soluble porphyrins, the separation of halogenated uncharged porphyrins followed by Ar-alkylation and metallation still appeared easier for the successful preparation of MnCl_xTE-2-PyP⁵⁺ (Scheme A) (Richards et al, Inorg. Chem. 35:1940 (1996); Kaufman et al, Inorg. Chem. 34:5073 (1995)):

Synthesis of H₂T-2-PyP β-chlorinated derivatives. β-Chlorination of H₂T-2-PyP was performed as described in the literature for H₂TPP analogues, using N-chlorosuccinimide (NCS) in chloroform under refluxing conditions (Ochsenbein et al, Angew. Chem. Int. Ed. Engl. 33:348 (1994)). The number of NCS equivalents used can be 4 or 6, depending on the degree of substitution desired (Table 3). The reaction can be followed by TLC (silica gel) using a mixture ethanol/dichloromethane (5:95) as eluant (Table 3 and Scheme B).

TABLE 3
H2ClxT-2-PyP (x = 1 to 4): R Soret band data and
yields with 4 and 6 equivalents of NCS.
	Yield (%)c
Porphyrin	Ra	λnm (ε/105 M−1cm−1)b	4 eq	6 eq
H2T-2-PyP	0.43	418.4
β-Cl1	0.47	419.6 (2.74)	30	—
β-Cl2a	0.50	421.4 (2.39)	10	5
β-Cl2b-2c	0.53	421.4 (2.62)	20	10
β-Cl3	0.55	422.8 (2.33)	10	15
β-Cl4	0.58	423.6 (2.13)	7	12
aTLC on silica with EtOH/CH2Cl2 (5:95) as eluant.
bin CHCl3 (estimated errors for ε are within 10%).
cin refluxing CHCl3 during 6 hours (c˜2 μM).

Each compound was purified by chromatography on silica gel (Wakogel C-300) using the same eluant. The structures of the main isomers were identified by mass spectrometry, and UV/VIS and ¹H-NMR spectroscopies (Table 3 and Scheme B). The bathochromic shift of the Soret band per chlorine on H₂T-2-PyP was only 1.3 nm compared to 3.5 nm reported previously for H₂TPP derivatives (Table 3) (Hoffmann et al, Bull. Soc. Chem. Fr. 129:85 (1992); Chorghade et al, Synthesis 1320 (1996); Wijesekera et al, Bull. Chem. Fr. 133:765 (1996)). Only one of the three dichlorinated regioisomers (β-Cl_2a derivative) was purified by chromatography on silica gel. Its two other regioisomers (β-Cl_2b and β-Cl_2c derivatives) exhibited the same R_f. Preliminary results showed that purification of H₂Br_xT-4-PyP (x=1 to 4) is more difficult. Indeed, using the same TLC system, β-Br₁ and β-Br_2a derivatives both have the same R_f, and no difference of R_f between β-Br_2b, β-Br_2c, β-Br₃ and β-Br₄ derivatives was observed, showing clearly that, in this case, R_f depends on the number of pyrroles β-substituted and not on the number of β-protons β-substituted.

¹H-NMR identification of H₂T-2-PyP β-chlorinated derivatives. ¹H-NMR allowed the identification of the products of the substitution reaction (Table 4 and FIG. 5). As described in the literature for H₂TPP analogues, the main regioisomer of H₂Cl₄T-2-PyP has chlorines in positions 7,8,17,18. Indeed, its ¹H-NMR spectrum shows an apparent singlet (doublet with J lower than 2 Hz), corresponding to four chemically equivalent β-protons coupled with the two pyrrolic protons which have lost their delocalization (Crossley et al, J. Chem. Soc., Chem. Commun. 1564 (1991). Nevertheless, another less polar fraction (R_f=0.60) was identified, according to its mass spectrum, as a mixture of other tetrachloro-regioisomers (¹H-NMR spectrum uninterpretable), representing approximately 50% by weight of both β-Cl₄ fractions, and showing that the β-substitution is only partially regioselective. According to the ¹H-NMR spectrum of the corresponding H₂Cl₃T-2-PyP⁵⁻ fraction, there are no apparent other regioisomers. The spectrum presents one singlet corresponding to the β-proton of the monoβ-substituted pyrrole and four doublets corresponding to the four β-protons of the two non-β-substituted pyrroles. Moreover, the asymmetry of this compound leads to a differentiation of the two NH protons. According to yields and ¹H-NMR spectra of H₂Cl_2aT-2-PyP (FIG. 5) and H₂Cl_2b+2cT-2-PyP, no predominant β-Cl₂ regioisomer was observed. Finally, the H₂C₁T-2-PyP spectrum shows one singlet and six doublets, but only one NH signal suggesting that in this case the asymmetry is too weak for the differentiation of the two NH protons.

TABLE 4
H2ClxT-2-PyP (x = 1 to 4): 1H-NMR
data (porphyrin ring) in CDCl3
	δppm(mult., Hz)a
	H2Cl1T-2-PyP	NH	−2.91(2H)
		CH	7.68(s, 1H)
			8.74(d, 1H, 5.5)
			8.76(d, 1H, 5.5)
			8.76(d, 1H, 6.0)
			8.88(d, 1H, 6.0)
			8.90(d, 1H, 6.0)
			8.94(d, 1H, 6.0)
	H2Cl2aT-2-PyP	NH	−2.98(2H)
		CH	8.70(s, 2H)
			8.82(d, 2H, 6.0)
			8.91(d, 2H, 6.0)
	H2Cl2bT-2-PyPb	NH	−3.04(2H)
		CH	8.59(s, 2H)
			8.78(d, 2H, 6.0)
			8.87(d, 2H, 6.0)
	H2Cl2cT-2-PyPb	NH	−2.84(1H)
			−2.87(1H)
		CH	8.61(s, 2H)
			8.73(d, 2H, <2.0)
			8.93(d, 2H, <2.0)
	H2Cl3T-2-PyP	NH	−3.08(1H)
			−3.15(1H)
		CH	8.56(s, 1H)
			8.72(d, 1H, 6.5)
			8.76(d, 1H, 6.5)
			8.82(d, 1H, 6.5)
			8.88(d, 1H, 6.5)
	H2Cl4T-2-PyP	NH	−3.14(2H)
		CH	8.74(d, 4H, <2.0)
	achemical shifts in ppm expressed relative to TMS by setting CDCl3 = 7.24 ppm.
	bone spectrum for the mixture of the two regioisomers (˜1:1 ratio).

N-ethylation and metallation. The N-ethylation of H₂T-2-PyP was efficiently accomplished using ethyl-p-toluenesulfonate, diethylsulfate or iodoethane as reagents, but the high toxicity of diethylsulfate and the low reactivity of iodoethane makes ethyl-p-toluenesulfonate (ETS) the best choice (Chen et al, J. Electroanal. Chem. 280:189 (1990); Kalyamasundaram, Inorg. Chem. 23:2453 (1984); Hambright et al, Inorg. Chem.: 15:1314 (1976); Alder et al, Chem. Brit. 14:324 (1978); Perree-Fauvet et al, 52:13569 (1996)). Some authors prefer performing N-alkylation after metallation in order to protect the pyrrole nitrogens (Perree-Fauvet et al, Tetrahedron 52:13569 (1996)). However, with direct treatment on the present free ligands, no N-ethylation of the pyrrole nitrogens was observed (subsequent metallation in aqueous solution was complete). The completion of ethylation as well as metallation can be followed by TLC (normal silica) using a highly polar eluant, a mixture of an aqueous solution of saturated potassium nitrate with acetonitrile (Batinic-Haberle et al, J. Biol. Chem. 273:24521 (1998)). The yields of this step (N-ethylation and metallation) were almost 100% (approximately 5% loss during the purification process). Since N-ethylation (or N-methylation) limits the free rotation of the pyridinium rings, each compound is in fact a mixture of four atropoisomers, and a further purification of each atropoisomer can be considered (Kaufmann et al, Inorg. Chem. 34:5073 (1995)). All the manganese porphyrins prepared had metal in the 3+state as demonstrated by the 20 nm hypsochromic shift of the Soret band (accompanied by the loss of splitting) upon the reduction of the metal-center by ascorbic acid.

Electrochemistry. The metal-entered redox behavior of all metalloporphyrin products was reversible. The half-wave potentials (E^O_1/2) were calculated as the average of the cathodic and anodic peaks and are given in mV vs NHE (Table 5). The average shift per chlorine is +55 mV (Table 5), which is in agreement with the values previously reported for H₂TPP derivatives (between +50 and +70 mV) (Sen et al, Chem. Soc. Faraday Trans. 93:4281 (1997); Autret et al, J. Chem. Soc. Dalton Trans. 2793 (1996); Hariprasad et al, J. Chem. Soc. Dalton Trans. 3429 (1996); Tagliatesta et al, Inorg. Chem. 35:5570 (1996); Ghosh, J. Am. Chem. Soc. 117:4691 (1995); Takeuchi et al, J. Am. Chem. Soc. 116:9730 (−1994); Binstead et al, Inorg. Chem. 30:1259-(1991); Giraudeau et al, J. Am. Chem. Soc. 101:3857 (1979)). This shift appears to be higher (˜+65 mV) between 0 and 1, and between 2 and 3 chlorines (Table 5). E^O_1/2 values of β-Cl_2a and the mixture β-Cl_2b+2c were not significantly different. The manganese redox state of MnCl₄TE-2-PyP⁵⁻(E⁰_1/2=+448 mV) and MnOBTMPyP⁴⁺(E^O_1/2=+480 mV) is 3+ and 2+, respectively. This difference may be explained by their difference in terms of redox potential (˜30 mV) but also by structural considerations, for instance an increased distortion of the porphyrin ring in the case of MnOBTMPyP⁴⁺. (Batinic-Haberle et al, Arch. Biochem. Biophys. 343:225 (1997); Ochsenbein et al, Angew. Chem. Int. Ed. Engl. 33:348 (1994)).

TABLE 5
MnClxTE-2-PyP5− (x = 1 to 4): Soret band data, redox potentials
and SOD activities.
	λnm (ε/		IC50/	kcat/
Mn-porphyrin	104 M−1 cm−1)a	Eo1/2 (Δ)b	10−9 Mc	107 M−1s−1
MnTE-2-PyP5−	453.8 (14.0)	+228 (71)	45	5.7
β-Cl1	455.6 (12.5)	+293 (65)	25	10
β-Cl2a	456.4 (10.6)	+342 (70)	20	13
β-Cl2b-2c	457.4 (11.2)	+344 (65)	20	13
β-Cl3	458.0 (9.5)	+408 (67)	10	26
β-Cl4	459.2 (8.0)	+448 (79)	6.5	40
MnTM-4-PyP5−		+060		0.4
MnTM-2-PyP5−		+220		6.0
MnOBTMPyP4−		+480		22
Cu,ZnSOD		+260		200
ain H2O (estimated errors for ε are within 10%).
bmV vs NHE, with estimated errors of 5 mV (Δ = peak to peak separation), and in the following conditions: 0.5 mM porphyrin, 0.1 M NaCl, 0.05 M phosphate buffer (pH 7.8).
cconcentration that causes 50% inhibition of cytochrome c reduction by O2− (estimated errors are within 10%).

Superoxide dismuting activities. SOD-like activities were measured as described previously, based on competition with cytochrome c (McCord et al, J. Biol. Chem. 244:6049 (1969)). MnCl_xTE-2-PyP⁵⁺ SOD-like activities are reported in Table 5, IC₅₀ (M) representing the concentration for one unit of activity (or the concentration that causes 50% inhibition of cytochrome c reduction by O₂⁻) and k_cat (M⁻¹s⁻¹) representing the rate constant for the superoxide dismutation reaction. The SOD-like activity per mole of MnCl₄TE-2-PyP⁵⁻ is approximately 2-, 7- and 100-fold higher than MnOBTMPyP⁴⁺, MnTM-2-PyP⁵⁺ and MnTM-4PyP⁵⁺, respectively (Faulkner et al, J. Biol. Chem. 269:23471 (1994); Batinic-Haberle et al, Arch. Biochem. Biophys. 343:225 (1997); Batinic-Haberle et al, J. Biol. Chem. 273:24521 (1998)). The SOD-like activity of MnCl₄TE-2-PyP⁵⁺ represents 20% of the activity of the Cu, Zn-SOD enzyme on a molar basis (40% per active site considering that the enzyme has two active sites) (Klug-Roth et al, J. Am. Chem. Soc. 95:2786 (1973)).

Test of stability. Each additional degree of chlorination increases the redox potential which is expected to be followed by the decrease in the pKa values of pyrrole nitrogens, as found for the series of meso-phenyl and meso-pyridyl β-substituted porphyrins as well as for β-substituted ones (Worthington et al, Inorg. Nucl. Chem. Lett. 16:441 (1980); Kadish et al, Inorg. Chem. 15:980 (1976)). The pKa, as a measure of the ligand-proton stability, is in turn a measure of the metal-ligand stability as well. Thus, the tetrachloro-compound is expected to be of decreased stability as compared to lesser chlorinated analogues. The stability of MnCl₄TE-2-PyP⁵⁺ was tested by measuring its SOD-like activity in the presence of excess EDTA. In the presence of a 10²-fold excess of EDTA, MnCl₄TE-2-PyP⁵⁻ (c=5×10⁻⁶ M) maintains its activity for sixteen hours (at 25° C.). A loss of activity (˜25%) was observed after forty hours, thus indicating the formation of some manganese—EDTA complex (K=10^14.05). These results confirm a relatively good stability of MnCl₄TE-2-PyP⁵⁻ when compared to MnOBTMPyP⁴⁺ (K=10^8.08) (Batinic-Haberle et al, Arch Biochem. Biophys. 343:225 (1997)).

Relationship between redox properties and SOD-like activities. The Cu, Zn-SOD enzyme is a dimer of two identical subunits, and thus has two active sites, which exhibit a redox potential close to the midpoint of the two half reaction values, as well as the same rate constants for each half reaction (Scheme C and Table 5) (Ellerby et al, J. am. Chem. Soc. 118:6556 (1996); Klug-Roth, J. Am. Chem. Soc. 95:2786 (1973)):

On the other hand, previous studies of O₂ dismutation catalyzed by MnTM-4-PyP⁵⁺ (E^O_1/2=+60 mV), using pulse radiolysis and stopped flow techniques, showed that the rate of the reduction of the metal by O₂⁻ is 10²-fold to 10³-fold lower than the rate of reoxidation of the metal (Faraggi, Oxygen Radicals in Chemistry and Biology, Bors et al (Eds): Walter de Gruyter and Co.; Berlin, Germany 1984, p. 419; Lee et al, J. Am. Chem. Soc. 120:6053 (1998)). Whereas a peak of SOD-like activity somewhere between +200 and +450 mV was first expected, plotting k_cat vs E^O_1/2 for MnCl_xTE-2-PyP⁵⁺ shows an exponential increase of the SOD-like activity, strongly suggesting that the limiting factor is still the reduction of the metal. This hypothesis however must be confirmed by measuring the rates of each half reaction as catalyzed by each. MnCl_xTE-2-PyP⁵⁻ compound. The relationship between activation free energy (ΔG^d) for superoxide dismutation and free energy change (ΔG^o) for MnCl_xTE-2-PyP⁵⁻ redox is linear (slope ˜+0.2), clearly showing the predominance of kinetic over thermodynamic factors in the theoretical optimal redox potential region (FIG. 6). According to this behavior, the activity of the Cu, Zn-SOD enzyme (k_cat=10⁹ M⁻¹s⁻¹ per active site) may be reached at approximately E^o_1/2=+570 mV (FIG. 3). However, due to both steric (distortion of the porphyrin ring) and thermodynamic factors, introducing a higher degree of β-chlorination is expected to stabilize the manganese in the 2+redox state, and thus, as in the case of MnOBTMPyP⁴⁺, limiting the rate of the reoxidation of the metal as well as inducing Mn (II) dissociation (Batinic-Haberle et al, Arch. Biochem. Biophys. 343:225 (1997); Ochsenbein et al, Angew. Chem. Int. Ed. Engl. 33:348 (1994)).

EXAMPLE VII

The ortho, meta and para isomers of manganese (II) 5,10,15,20-tetrakis(N-methylpyridyl)porphyrin, MnTM-2-PyP⁵⁺, MnTM-3-PyP⁵⁺, and MnTM-4-PyP⁵⁺, respectively, were analyzed in terms of their superoxide dismutase (SOD) activity in vitro and in vivo. The impact of their interaction with DNA and RNA on the SOD activity in vivo and in vitro was also analyzed. Differences in their behavior are due to the combined steric and electrostatic factors. In vitro catalytic activities are closely related to their redox potentials. The half-wave potentials (E_1/2) are +0.220 mV, +0.052 mV and +0.060 V vs normal hydrogen electrode (NHE), while the rates of dismutation (k_cat), are 6.0×10⁷ M⁻¹ s⁻¹, 4.1×10⁶ M⁻¹ s⁻¹ and 3.8×10⁶ M⁻¹ s⁻¹ for the ortho, meta and para isomers, respectively.

However, the in vitro activity is not a sufficient predictor of in vivo efficacy. The ortho and meta isomers, although of significantly different in vitro SOD activities, have fairly close in vivo SOD efficacy due to their similarly weak interactions with DNA. In contrast, due to a higher degree of interaction with DNA, the para isomer inhibited growth of SOD-deficient Escherichia coli. For details, see Batinic-Haberle et al, J. Biol. Chem. 273(38):24521-8 (Sep. 18, 1998).

EXAMPLE VIII

Metalloporphyrins are Potent Inhibitors of Lipid Peroxidation

Materials and Methods

L-Ascorbic acid, n-butanol, butylated hydroxytoluene, cobalt chloride, iron (II) chloride, phosphoric acid (85%), sodium hydroxide, potassium phosphate, tetrabutylammonium chloride, and 1,1,3,3-tetramethoxypropane were purchased from Sigma (St. Louis, Mo.). Acetone, concentrated hydrochloric acid, 4,6-dihydroxy-2-mercaptopyrimidine (thiobarbituric acid), NH₄PF₆, zinc chloride, 5,10, 15, 20-tetrakis (4-benzoic acid) porphyrin (H₂TBAP)*, 5,10,15,20-tetrakis (N-methylpyridinium-4-yl) porphyrin (H₂TM-4-PyP), and Trolox were purchased from Aldrich (Milwaukee, Wis.). Ferric 5,10,15,20-tetrakis (4-benzoic acid) porphyrin (FeTBAP) was purchased from Porphyrin Products (Logan, Utah). 5,10,15,20-tetrakis (N-methylpyridinium-2-yl) porphyrin (H₂TM-2-PyP) was purchased from MidCentury Chemicals (Posen, Ill.). (+)-Rutin was purchased from Calbiochem (La Jolla, Calif.). Manganese chloride was purchased from Fisher (Fair Lawn, N.J.) and ethanol USP was purchased from AAPER
*Also known as 5,10,15,20-tetrakis (4-carboxyphenyl) porphyrin (H₂TCPP) Alcohol and Chemical Co. (Shelbville, Ky.). All solutions were prepared in Milli-Q Plus PF water (Millipore, Bedford, Mass.).

Preparation and Analysis of Metalloporphyrins

The metalloporphyrins MnTBAP, CoTBAP and ZnTBAP were made using methods described previously (Day et al, J. Pharmacol. Exp. Ther. 275:1227 (1995)). MnTM-4-PyP, CoTM-4-PyP and ZnTM-4-PyP were synthesized by the following method. A 1.5 molar excess of manganese, cobalt or zinc chloride was mixed with H₂TM-4-PyP that was dissolved in de-ionized water. The reaction mixture was heated to 80° C. and metal ligation was followed spectrophotometrically (UV-2401PC, Shimadzu, Columbia, Md.). Excess metal was removed by passing the mixture through a column containing Bio-Gel P-2 (BioRad, Richmond, Calif.) that selectively retained MnTM-4-PyP. MnTM-4-PyP was eluted with 0.01 N HCl after extensive washing of the column with water. MnTM-4-PyP, CoTM-4-PyP and ZnTM-4-PyP were characterized in terms of their reported Soret bands. The Soret band for MnTM-4-PyP is at 463 nm with an extinction coefficient of (ε)=1.3×10⁵ M⁻¹ cm⁻¹, the Soret band for ZnTM-4-PyP is at 437 nm with an extinction coefficient of (ε)=2.0×10⁵ M⁻¹ cm⁻¹ (Pasternack et al, Inorg. Chem. 12:2606 (1973)) and the Soret band for CoTM-4-PyP is at 434 nm with an extinction coefficient of (E)=2.15×10⁵ M⁻¹ cm⁻¹ (Pasternack et al, Biochemistry 22:2406 (1983)). Manganese β-octabromo-meso-tetrakis-(N-methylpyridiniumyl) porphyrin. (MnOBTM-4-PyP) was synthesized as previously described (Batinic-Haberle et al, Arch. Biochem. Biophys. 343:225 (1997)) and has a Soret band at 490 nm with an extinction coefficient (ε)=8.56×10⁴ M⁻¹ cm⁻¹. H₂TM-2-PyP was metallated with a 1:20 porphyrin to manganese ratio in water (pH>11) at room temperature. Upon completion of metallation, MnTM-2-PyP was precipitated by the addition of a concentrated aqueous solution of NH₄ PF₆. The precipitate was washed with 2-propanol:diethyl ether (1:1) and dried in vacuo at room temperature. The PF₆⁻ salt of MnTM-2-PyP was dissolved in acetone, filtered and a concentrated acetone solution of tetrabutylammonium chloride was added until the porphyrin had precipitated as its chloride salt. The precipitate was washed with acetone and dried in vacuo at room temperature. The Soret band for MnTM-2-PyP was found at 453 nm with an extinction coefficient (ε)=1.29×10⁵ M⁻¹ cm⁻¹.

Preparation of Rat Brain Homogenates

Frozen adult Sprague-Dawley rat brains (Pel-Freez, Rogers, A R) were homogenized with a polytron (Turrax T25, Germany) in 5 volumes of ice cold 50 mM potassium phosphate at pH 7.4. Homogenate protein concentration was determined with the Coomassie Plus protein assay (Pierce, Rockford, Ill.) using bovine serum albumin as a standard. The homogenate volume was adjusted with buffer to give a final protein concentration of 10 mg/ml and frozen as aliquots at −80° C.

Oxidation of Rat Brain Homogenates

Rat brain homogenates (2 mg protein) were incubated with varying concentrations of antioxidant at 37° C. for 15 minutes. Oxidation of the rat brain homogenate was initiated by the addition of 0.1 ml of a freshly prepared anaerobic stock solution containing iron (II) chloride (0.25 mM) and ascorbate (1 mM) as previously reported (Braughler et al, J. Biol. Chem. to 262:10438 (1987)). Samples (final volume 1 ml) were placed in a shaking water bath at 37° C. for 30 minutes. The reactions were stopped by the addition of 0.1 ml of a stock butylated hydroxytoluene (60 mM) solution in ethanol.

Lipid Peroxidation Measurement

The concentration of thiobarbituric acid reactive species (TBARS) in rat brain homogenates was used as a index of lipid peroxidation (Bernhem et al, J. Biol. Chem. 174:257 (1948); Witz et al, J. Free Rad. Biol. Med. 2:33 (1986); Kikugawa et al, Anal. Biochem. 202:249 (1992); Jentzsch et al, Free Rad. Biol. Med. 20P251 (1996)). Malondialdehyde standards were obtained by adding 8.2 μl of 1,1,3,3-tetramethoxypropane in 10 ml of 0.01 M HCl and mixing for 10 minutes at room temperature. This stock was further diluted in water to give standards that ranged from 0.25 to 25 μM. Samples or standards (200 μl) were acidified with 200 μl of 0.2 M phosphoric acid in 1.5 ml locking microfuge tubes. The color reaction was initiated by the addition of 25 μl of a 0.11M thiobarbituric acid solution and samples were placed in a 90° C. heating block for 45 minutes. TBARS were extracted with 0.5 ml of n-butanol by vortexing samples for 3 minute and chilling on ice for 1 minute. The samples were then centrifuged at 12,000×g for 3 minutes, 150 μl aliquots of the n-butanol phase were placed in each well of a 96-well plate and read at 535 nm in a Thermomax platereader (Molecular Devices, Sunnyvale, Calif.) at 25° C. Sample absorbencies were converted to MDA equivalencies (μM) by extrapolation from the MDA standard curve. None of the antioxidants at concentrations employed in these studies affected the reaction of MDA standards with thiobarbituric is acid and reactions without TBA were used as subtraction blanks.

Statistical Analyses

Data were presented as their means±SE. The inhibitory concentration of antioxidants that decreased the degree of lipid peroxidation by 50% (IC₅₀) and respective 95% confidence intervals (Cl) were determined by fitting a sigmoidal curve with variable slope to the data (GraphPad Prizm, San Diego, Calif.).

Results

Comparison of metalloporphyrins with other antioxidants in iron/ascorbate-mediated lipid peroxidation.

The objective of these studies was to investigate whether metalloporphyrins could inhibit lipid peroxidation and to compare their to potencies with those of previously characterized antioxidants that include enzymatic antioxidants (SOD and catalase) and non-enzymatic antioxidants (water soluble vitamin E analog, trolox, and plant polyphenolic flavonoid, rutin) (FIG. 7). The time course of lipid peroxidation was determined in rat brain homogenates using iron and ascorbate as initiators of lipid oxidation and the formation of thiobarbituric reactive species (TBARS) as an index of lipid peroxidation. A linear increase in the formation of TBARS occurred between 15 to 90 minutes of incubation at 37° C. (FIG. 8). Based on this result, an incubation time of 30 minutes was selected to test the ability of metalloporphyrins and other antioxidants to inhibit lipid peroxidation. (FIG. 9). Of the agents tested, the manganese porphyrins that have the highest SOD activities, MnOBTM-4-PyP and MnTM-2-PyP, were found to be the most potent lipid peroxidation inhibitors with calculated IC₅₀s of 1.3 and 1.0 μM respectively. (Table 6). Bovine CuZnSOD was moderately active with a calculated IC₅₀ of 15 μM while trolox and rutin were much less potent with calculated IC₅₀s of 204 and 112 μM, respectively. In this system, catalase (up to concentrations of 1 mg/ml) did not inhibit iron/ascorbate-initiated lipid peroxidation.

TABLE 6
Comparison of Antioxidant Properties
	SOD	Redox Potential	Lipid Peroxidationc
Antioxidants	(U/mg)a	(E1/2, V)b	IC50 [μM]	95% Cl [μM]
CuZnSOD	5,100	+0.35	15	13-17
Trolox	—	—	204	135-308
Rutin	—	—	113	99-129
MnTM-2-PyP	8,500	+0.22	1.0	0.4-2.2
MnOBTM-4-PyP	18,460	+0.48	1.3	0.8-2.2
MnTM-4-PyP	547	+0.06	16	12-22
MnTBAP	179	−0.19	29	23-37
CoTM-4-PyP	113	+0.42	17	14-22
CoTBAP	24	+0.20	21	13-33
FeTBAP	24	+0.01	212	144-311
ZnTM-4-PyP	trace	—	241	159-364
ZnTM-2-PyP	trace	—	591	423-827
ZnTBAP	trace	—	843	428-1660
aUnit of SOD activity defined as the amount of compound that inhibits one half the reduction of cytochrome c or photoreduction of NBT.
bMetal centered redox potentials vs NHE (Mn+3/Mn+2; Co+3/Co+2; Fe+3/Fe+2). If not otherwise specified, E1/2 were obtained at pH 7.8.
cThe amount of thiobarbaturic acid reactive substances produced in a rat brain homogenate by 30 minutes of incubation of iron and ascorbate.

Effect of Different Metal Chelates on the Ability of Porphyrins to Inhibit Lipid Peroxidation.

A wide range of metals can be covalently ligated by porphyrins and that confers different redox potentials and SOD activities (Table 6). The ability of different metal chelates to influence a porphyrin's ability to inhibit lipid peroxidation was tested. Several different metal analogs of TBAP were examined in the iron/ascorbate-initiated lipid peroxidation model (FIG. 10). Both the manganese and cobalt TBAP analogs had similar efficacy with calculated IC₅₀ of 29 and 21 μM, respectively. The FeTBAP analog was an order of magnitude less potent with a calculated IC₅₀ of 212 μM. The ZnTBAP analog was much less active than the other metal analogs with a calculated IC₅₀ of 946 μM. This potency difference between the zinc and the other metals reflects the importance of metal centered verses ring structure redox chemistry since zinc can not readily change its valence. The ranked potencies of tested metalloporphyrins based on IC₅₀s were as follows: MnTM-2-PyP═MnOBTM-4-PyP>MnTM-4-PyP═CoTM-4-PyP>CoTBAP═MnTBAP>FeTBAP═ZnTM-4-PyP>ZnTM-2-PyP>ZnTBAP.

Comparison of a Series of Tetrakis N-Methylpyridyl Porphyrin (TMPyP) Analogs as Inhibitors of Lipid Peroxidation.

Recently, several manganese analogs of N-methylpyridyl porphyrins have been found to possess large differences in SOD activities (Table 6). MnTM-2-PyP and MnTM-4PyP differ structurally with respect to the position of the N-methylpyridyl group to the porphyrin ring (ortho vs para) as well as in SOD activity by a factor of six. Substitution of zinc in these porphyrin analogs results in loss of SOD activity. These TMPyP analogs were compared for their ability to inhibit lipid peroxidation (FIG. 11). The movement of the N-methylpyridyl group from the para- to the ortho-position in the manganese porphyrin resulted in a 15-fold increase in potency. Since MnTM-2-PyP possesses a more positive redox potential than MnTM-4-PyP (+0.22 vs +0.06, respectively), this data suggests that both the redox potential and the related SOD activity may contribute to the increased potency of the MnTM-2-PyP analog.

EXAMPLE IX

Demonstration That Mn TE-2-PyP Can Be Effectively Used to Attenuate Oxidant Stress Mediated Tissue Injury

The ability of Mn TE-2-PyP to attenuate injury associated with 60 minutes of global ischemia followed by 90 minutes of reperfusion was assessed in an isolated, perfused mouse liver model. Excised livers were perfused with a buffered salt solution for 15 minutes after which the metalloporphyrin was introduced into the perfusate and the liver perfused in a recirculating system for ar additional 15 minutes. The livers were then rendered globally ischemic under normal thermic conditions for 60 minutes. Following the ischemic period the livers were perfused for 90 minutes with perfusate supplemented with 10 μm Mn TE-2-PyP. In this model the ischemia/reperfused livers have a marked release of hepatocellular enzymes, aspartate transaminase, alanine transaminase, and lactate dehydrogenase during the first 2½ minutes of reperfusion. This is followed by a progressive release of hepatocellular enzymes indicating hepatocellular injury over the 90 minute perfusion period. Administration of Mn TE-2-PyP was highly efficacious in attenuating the liver injury, blocking virtually all of the acute hepatocellular enzyme release and blocking progressive hepatocellular enzyme release over the 90 minute perfusion period. At the end of the experiments liver is treated with the metalloporphyrin. It has demonstrated excellent oxygen consumption and a normal perfusion pattern. They remain firm and with a normal texture to gross morphologic examination. Livers with no drug treatment did not consume oxygen normally and became edematous, soft, and had a mottled appearance consistent with poor perfusion.

EXAMPLE X

Effects of Mn TNI-2-PyP on Vascular Tone

Rats were anesthetized and a femoral vein and carotid artery were cannulated. While blood pressure was monitored by the carotid artery, Mn TM-2-PyP was injected i.v. at doses ranging from 0.1 to 3.0 mg/kg. Mean arterial pressure fell from 100-125 mmHg to 50-60 mmHg within five to ten minutes. The effect was transient, lasting up to 30 minutes at doses of 0.1 to 0.25 mg/kg. At doses of 1-3 mg/kg the effect was prolonged, lasting up to two hours. The effect can be blocked by administration of inhibitors of nitric oxide synthase demonstrating that the role of Mn TM-2-PyP is being modulated by nitric oxide. Scavenging of superoxide in vascular walls would potentiate the effects of nitric oxide producing hypotension.

EXAMPLE XI

Regulation of Airway Reactivity Using Nin TM-2-PyP

Mice were sensitized by intraperitoneal injection of ovalbumin twice, 14 days apart. Fourteen days after the second i.p. injection they were challenged with aerosolized ovalbumin daily for three days. Forty-eight hours after the third inhalation of ovalbumin they were given a 1 minute methacholine challenge and airway hyperreactivity followed using a Buxco body plethysmograph. Significant increases in airway resistance as measured by the PENH index occurred at doses of 20, 30 and 40 mg/ml of methacholine. At all doses of methacholine prior intratracheal instillations of 2 μg Mn TM-2-PyP given daily for 4 days resulted in a statistically significant reduction in the airway hyperreactivity. This dose of Mn TM-2-PyP is equivalent to 0.8 mg/kg whole body dose.

EXAMPLE XII

Treatment of Bronchopulmonary Dysplasia Using Mn TE-2-PyP

Neonatal baboons were delivered prematurely by Caesarian section and then treated either with 100% oxygen or only sufficient PRN FIO₂ to maintain adequate arterial oxygenation. To establish the model, thirteen 100% oxygen treated animals and twelve PRN control animals were studied. Treatment with 100% oxygen results in extensive lung injury manifested by days 9 or 10 of exposure and characterized by delayed alveolarization, lung parenchymal inflammation, and poor oxygenation. This is characteristic of the human disease, bronchopulmonary dysplasia and is thought to be mediated, at least in part, by oxidative stress on the developing neonatal lung. In a first trial of Mn TE-2-PyP, a neonatal baboon was delivered at 140 days gestation and placed in 100% oxygen. The animal received 0.5 mg/kg/24 hr Mn TE-2-PyP qd given i.v. in a continuous infusion over the entire 10 day study period. This animal showed marked improvement of the oxygenation index. There was no evidence of clinical decompensation of the lungs at days 9 and 10. Lung pathology demonstrated absence of inflammation and a marked decrease in the lung injury found in the prior animals treated with 100% oxygen under identical conditions. This suggests that Mn TE-2-PyP can be used to treat oxidant stress in the premature newborn.

All documents cited above are hereby incorporated in their entirety by reference. Appln. No. 60/064,116, filed Nov. 3, 1997, is also incorporated in its entirety by reference.

One skilled in the art will appreciate from a reading of this disclosure that various changes in form and detail can be made without departing from the true scope of the invention.

Claims

1. A compound of formula or pharmaceutically acceptable salt thereof,

wherein each R is, independently, a C1-C8 alkyl group, and each P is, independently, an electron withdrawing group or hydrogen,

wherein when each R is methyl and each P is hydrogen, said compound is completed with a metal selected from the group consisting of manganese, iron, copper, cobalt, nickel or zinc.

2. The compound according to claim 1 where each R is independently a C1-C4 alkyl group.

3. The compound according to claim 2 wherein each R is, independently, a methyl, ethyl or isopropyl group.

4. The compound according to claim 3 wherein each R is, independently, a methyl or an ethyl group.

5. The compound according to claim 1 wherein each P is, independently, hydrogen or an electron withdrawing group selected from the group consisting of —NO2, a halogen, a nitrile, a vinyl group and a formyl group.

6. The compound according to claim 1 wherein at least one P is a halogen.

7. The compound according to claim 1 wherein one or two P's are formyl groups and the remaining P's are hydrogen.

8. The compound according to claim 1 wherein one P is a formyl group and the remaining P's are hydrogen.

9. The compound according to claim 1 wherein one or two P's are —NO2 and the remaining P's are hydrogen.

10. The compound according to any one of claims 1-9 wherein said compound is complexed with a metal selected from the group consisting of manganese, iron, copper, cobalt, nickel or zinc.

11. The compound according to claim 10 wherein said compound is complexed with manganese.

12. The compound according to claim 1 wherein each R is a methyl or ethyl group, each P is a hydrogen, and said compound is complexed with manganese.

13. The compound accord ng to claim 1 wherein each R is a methyl or ethyl group, at least one P is Br and the remaining P's are hydrogen and said compound is complexed with manganese.

14. The compound according to claim 1 wherein said compound is a mixture of atropoisomers αααα, αααβ, ααββ and αβαβ.

15. The compound according to claim 1 wherein said compound is a mixture of αααβ and αααα atropoisomers.

16. A method of protecting cells from oxidant-induced toxicity comprising contacting said cells with a protective amount of a compound of formula or pharmaceutically acceptable salt thereof,

wherein each R is, independently, a C1-C8 alkyl group, and each P is, independently, an electron withdrawing group or hydrogen.

17. The method according to claim 16 wherein said compound is complexed with a metal selected from the group consisting of manganese, iron, copper, cobalt, nickel or zinc.

18. The method according to claim 16 wherein said cells are mammalian cells.

19. A method of treating a pathological condition of a patient resulting from oxidant-induced toxicity comprising administering to said patient an effective amount of a compound of formula or pharmaceutically acceptable salt thereof,

wherein each R is, independently, a C1-C8 alkyl group, and each P is, independently, an electron withdrawing group or hydrogen.

20. The method according to claim 19 wherein said compound is complexed with a metal selected from the group consisting of manganese, iron, copper, cobalt, nickel or zinc.

21. A method of treating a pathological condition of a patient resulting from degradation of NO. or a biologically active form thereof, comprising administering to said patient an effective amount of a compound of formula or pharmaceutically acceptable salt thereof,

wherein each R is, independently, a C1-C8 alkyl group, and each P is, independently, an electron withdrawing group or hydrogen.

22. The method according to claim 21 wherein said compound is complexed with a metal selected from the group consisting of manganese, iron, copper, cobalt, nickel or zinc.

23. A method of treating a patient for inflammatory lung disease comprising administering to said patient an effective amount of a compound of formula or pharmaceutically acceptable salt thereof,

wherein each R is, independently, a C1-C8 alkyl group, and each P is, independently, an electron withdrawing group or hydrogen.

24. The method according to claim 23 wherein said compound is complexed with a metal selected from the group consisting of manganese, iron, copper, cobalt, nickel or zinc.

25. The method according to claim 24 wherein said metal is manganese.

26. The method according to claim 23 wherein said inflammatory lung disease is a hyper-reactive airway disease.

27. The method according to claim 23 wherein said inflammatory lung-disease is asthma.