The focus in this review is directed to molecular aspects of aluminum toxicity in animal and plant cells. Unique thermodynamic features of Al(lII) ions impart biological specificity which may form the biochemical basis of aluminum interactions with cellular constituents. Current knowledge about aluminum‐specific, molecular interactions is rather scanty. Al(III) ions may coordinate with nucleotides or complex to phospholipids resulting in impaired genetic expression or aluminum‐induced membrane phase changes, respectively. Aluminum‐specific structures are formed with proteins involved in iron transport. and with calmodulin, a key regulatory protein involved in coordinating the actions of second messengers. A novel hypothesis is put forward that Al‐calmodulin may be a primary lesion that occurs in the broadly defined syndrome of aluminum toxicity. Pertinent research areas are discussed.

Molecular aspects of aluminum toxicity

Since uranium has several common paramagnetic oxidation states, i.e., 5f 1 U, 5f 2 U, and 5f 3 U, one of the analytical methods commonly used for characterizing uranium complexes is measurement of magnetic susceptibilities. A variety of solid state and solution methods have been used to obtain magnetic data on uranium complexes under different conditions at both room and low temperature, and data are often cited to support the assignment of oxidation state. Statements like “the magnetic moment of compound X is consistent with U” are common in uranium papers. However, it has been pointed out repeatedly in the literature that evaluating the magnetic susceptibility of uranium complexes is not as straightforward as it is for transition metal and lanthanide complexes. For any metal complex, the magnetic moment arises from the sum of the contributions of the ground state and the low-lying thermally accessible excited states as well as any temperature-independent magnetism in the system. For most 4f lanthanides, the ground state is defined by a LJ Russell−Saunders term because spin−orbit coupling is large and ligand field splitting is small due to the limited radial extension of the 4f orbitals. Roomtemperature magnetic moments can be approximated for most of the lanthanides by μJ = g[J(J + 1)] 1/2 and correlated with oxidation state. This approximation does not apply to Sm and Eu because they have low-lying excited states, but the range of their experimental values is well defined. The situation is more complicated for d transition metals where the ligand field effects are strong and can diminish the contribution of the orbital angular momentum to the magnetic moment (orbital quenching). However, an operational solution exists for transition metal complexes: a spin-only approximation, μS = 2[S(S + 1)] , can be used as a starting point and then modified for orbital contributions depending on the d configuration and symmetry. The actinides are intermediate between lanthanides and transition metals in the sense that the greater radial extension of the 5f orbitals means that ligand field effects cannot be ignored, yet spin−orbit coupling is large. In a case where two effects can be comparable, finding a simple model is usually more difficult. Neither the spin-only μS approximation used for transition metals nor the total angular momentum μJ approximation used for the lanthanides is adequate for uranium, although μJ is often used as an approximate starting point. A detailed theoretical analysis on actinide magnetism has recently been published that defines these issues more formally. A further complication for uranium is that the μJ values of 3.62 and 3.58 μB often used to approximate magnetism for U 3+ and U, respectively, are similar enough to be within the experimental error of typical measurements. It is therefore difficult to differentiate U and U from room-temperature magnetic moments. As a consequence, it is preferable to obtain variable-temperature data with a superconducting quantum interference device (SQUID) or vibrating sample magnetometer (VSM) since the magnetic moment of a 5f 2 U ion typically approaches zero at low temperature due to a singlet ground state. This does not occur with half-integral 5f 3 U and 5f 1 U. Unfortunately, although variable-temperature analysis is the state-of-the-art method to evaluate the magnetism of uranium complexes, such data are not always easily obtained by all research groups and even the shapes of variable-temperature plots are not always consistent for a given ion. Despite these difficulties, many claims of oxidation state based on room-temperature magnetic moments are in the literature. These claims often cite other papers with similar magnetic moments to support the oxidation state assignment without mentioning the papers in which different magnetic moments were observed for that oxidation state. To our knowledge, no comprehensive collection of magnetic data on uranium complexes has been assembled to survey the actual range of data in the literature. To remedy this deficiency, the data in this review were compiled. It was of interest to determine if there were any trends in the ranges of magnetic moments reported for uranium complexes and the frequency of typically observed values. Magnetic data on over 500 complexes of uranium in the common +3, +4, and +5 oxidation states as

Magnetic Susceptibility of Uranium Complexes

Oxovanadium(IV) and oxovanadium(V) complexes relevant to biological systems.

Highly Enantioselective Catalytic Conjugate Addition of N‐Heterocycles to α,β‐Unsaturated Ketones and Imides

Hexavalent chromium (Cr (VI)) is reduced intracellularly to Cr (V), Cr (IV) and Cr (HI) by ascorbate (Asc), cysteine and glutathione (GSH). These metabolites induce a spectrum of genomic DNA damage resulting in the inhibition of DNA replication. Our previous studies have shown that treatment of DNA with Cr (III) or Cr (VI) plus Asc results in the formation of DNA-CrDNA crosslinks (Cr-DDC) and guanine-specific arrests of both prokaryotic and mammalian DNA polymerases. GSH not only acts as a reductant of Cr (VI) but also becomes crosslinked to DNA by Cr, thus, the focus of the present study was to examine the role of GSH in Cr-induced DNA damage and polymerase arrests. Co-incubation of Cr (III) with plasmid DNA in the presence of GSH led to the crosslinking of GSH to DNA. GSH co-treatment with Cr (III) also led to a decrease in the degree of Cr-induced DNA interstrand crosslinks relative to Cr(III)alone, without affecting total Cr DNA binding. DNA polymerase arrests were observed following treatment of DNA with Cr (III) alone, but were markedly reduced when GSH was added to the reaction mixture. Pre-formed polymerase-arresting lesions (Cr-DDC) were not removed by subsequent addition of GSH. Treatment of DNA with Cr (VI), in the presence of GSH, resulted in crosslinking of GSH to DNA, but failed to produce detectable DNA interstrand crosslinks or polymerase arrests. The inhibitory effect of GSH on Cr-induced polymerase arrest was further confirmed in human genomic DNA using quantitative PCR (QPCR) analysis. Treatment of genomic DNA with Cr (III) resulted in a marked inhibition of the amplification of a 1.6 kb target fragment of the p53 gene byTagpolymerase. This was almost completely prevented by co-treatment with GSH and Cr (III). These results indicate that Cr-induced DNA interstrand crosslinks, and not DNA-Cr-GSH crosslinks, are the principal lesions responsible for blocking DNA replication. Moreover, the formation of DNA-Cr-GSH crosslinks may actually preclude the formation of the polymerase arresting lesions. (Mol Cell Biochem222:173-182, 2001)

Linda Mattucci

Papers

Xanthine, hypoxanthine and guanine copper(II) complexes

Guanine complexes with first row transition metal perchlorates

Copper(II) perchlorate complexes with N-methyl substituted xanthines

Aluminum chloride complexes with purine, adenine and guanine

Adducts of guanine with chromium(III), iron(III) and oxovanadium(IV) chlorides