Chemistry of bleomycin. XVI epi-bleomycin

N-methyl threonine analogues of deglycobleomycin A2: synthesis and evaluation

The synthesis of 5 and its D-allo-threonine epimer 6 and the comparison of their DNA cleavage efficiency and selectivity with that of deglycobleomycin A2 (3) are detailed. The studies illustrate that N-methylation of the L-threonine subunit within deglycobleomycin A2 dramatically reduces the DNA cleavage efficiency (10-15-fold), weakens and nearly abolishes the inherent DNA cleavage selectivity, but has little effect on the inherent oxidation capabilities of the activated Fe(III) complexes. The results are consistent with a previously unrecognized prominent role for the threonine NH and the potential importance of a hydrogen bond to the Fe(III) hydroperoxide complex of bleomycin or a subsequent activated complex implicated in recent structural models.

Synthesis and evaluation of deglycobleomycin A2 analogues containing a tertiary N-methyl amide and simple ester replacement for the L-histidine secondary amide: direct functional characterization of the requirement for secondary amide metal complexation

The synthesis and comparative examination of 3-5, analogues of deglycobleomycin A2 (2) which address the inferred importance of the L-histidine secondary amide directly, are detailed. The agent 3 lacks only the L-histidine beta-hydroxy group of deglycobleomycin A2 and the corresponding agents 4 and 5 incorporate a tertiary N-methyl amide and simple ester in place of the L-histidine secondary amide. The DNA cleavage properties of 3 proved essentially indistinguishable from those of deglycobleomycin A2 (2) confirming that the distinctions between bleomycin A2 (1) and deglycobleomycin (2) are due to the removal of the disaccharide and not the introduction of the L-histidine free beta-hydroxy group. The agents 4 and 5 containing a tertiary N-methyl amide and ester in place of the L-histidine secondary amide were found to cleave duplex DNA but to do so in a nonsequence selective fashion with a substantially reduced efficiency and a diminished double to single strand cleavage ratio that are only slightly greater than that of free iron itself. These latter observations establish the functional requirement for the L-histidine secondary amide and are consistent with the proposals that the L-histidine deprotonated secondary amide is required for functional metal chelation and activity.

Synthesis and evaluation of potential N pi and N sigma metal chelation sites within the beta-hydroxy-L-histidine subunit of bleomycin A2: functional characterization of imidazole N pi metal complexation

The synthesis and evaluation of 4 and 5, fully functionalized deglycobleomycin A2 (2) analogues incorporating an oxazole and a pyrrole in place of the beta-hydroxy-L-histidine imidazole, are detailed. The oxazole agent is only capable of Npi metal complexation through a form related to the N1-H imidazole tautomer of bleomycin A2 (1) while the pyrrole agent may potentially mimic the Nsigma metal complexation capabilities of the imidazole N3-H tautomer. Metal complexes (Fe-II, Fe-III) of 4 and 5 were found to cleave duplex DNA in the presence of O2 (Fe-II) or H2O2 (Fe-III). The oxazole agent 4 which is incapable of Nsigma metal chelation was found to behave analogous to, albeit slightly less effectively than, deglycobleomycin A2 resulting in the characteristic 5'-GC/5'-GT sequence selective cleavage of duplex DNA directly confirming that imidazole/oxazole Npi metal chelation is sufficient for functional reactivity. Importantly, the effective substitution of the oxazole O-1 for the histidine N-1 further illustrates that this group does not require deprotonation upon metal complexation, oxygen activation, or the ensuing oxidation reactions, that the functional bleomycin A2 tautomer is the imidazole N'-H tautomer, and that the imidazole N'-H functionality is not contributing to the polynucleotide recognition through H-bonding to the phosphate backbone or nucleotide bases. In contrast, the pyrrole agent 5 which is incapable of Npi metal chelation, but possesses the capabilities of functioning as a Nsigma metal donor was also found to cleave duplex DNA, but does so in a nonsequence selective fashion with a significantly reduced efficiency and a diminished double to single strand cleavage ratio both only slightly above that of background iron itself. These observations are analogous to those made with 3 which lacks the imidazole altogether and further support the observations that Nsigma coordination, not Npi coordination, of the imidazole is required for the functional activity of bleomycin A2.

Synthesis of key analogs of bleomycin A2 that permit a systematic evaluation of the linker region: identification of an exceptionally prominent role for the L-threonine substituent

The synthesis of a full series of analogs 2b-k of deglycobleomycin A2 (2a) containing systematic variations in the linker domain of bleomycin A2 (1) is described. The agents 2b-k, which are not accessible through structural modification of 1 or 2a, constitute key substructure analogs incorporating deep-seated structural modifications in the linker domain capable of delineating the contribution of the individual backbone substituents to the DNA cleavage efficiency, characteristic DNA cleavage selectivity, and double strand to single strand DNA cleavage ratio. The comparative examination of the DNA cleavage properties of the Fe(II) and Fe(III) complexes of 2a-k upon activation by O2-thiol or H2O2, respectively, revealed several characteristic features and trends. First, none of the substituents affect the characteristic 5'-GC, 5'-GT > 5'-GA DNA cleavage selectivity of bleomycin A2. In contrast, an exceptionally prominent role for the L-threonine substituent and an important role for the C4-methyl substituent of the (2S,3S,4R)-4-amino-3-hydroxy-2-methylpentanoic acid subunit were observed on the DNA cleavage efficiency of the agents. Similarly, the L-threonine substituent was found to substantially increase the ratio of double strand to single strand DNA cleavage events (2-3 times). In a w794 DNA cleavage assay, shortening the linker region by two carbons resulted in an exceptionally large reduction in DNA cleavage efficiency (125 times) and provided an agent that was only 1.3 times more effective than Fe(III) indicating that this deep-seated modification essentially destroys the DNA cleavage capabilities of the agent. The L-threonine substituent contributes in an exceptional manner, and its removal resulted in a 25 times reduction in DNA cleavage efficiency. A substantial contribution was observed for the C4-methyl group on the 4-aminobutanoic acid subunit and its removal resulted in a 7 times reduction in DNA cleavage efficiency. Little effect for the C3-hydroxyl and C2-methyl substituents on the 4- aminobutanoic acid subunit was observed (0-2.5 times) and even their inversion of stereochemistry had little impact on DNA cleavage efficiency or selectivity. Notably, the magnitude of the previously unappreciated L-threonine substituent contribution to the DNA cleavage efficiency and on the ratio of double to single strand DNA cleavage events is the largest effect observed to date including the well recognized disaccharide potentiation (6 times) of the DNA cleavage properties. Consequently, the past role and relative importance of the L-threonine subunit and substituent has been underestimated. Moreover, the cumulative effect of the two important linker chain substituents clearly illustrate that the functional role of this domain is much more important than its simply serving as a linker.

DNA damage and growth inhibition in cultured human cells by bleomycin congeners

Bleomycin is hypothesized to cause cell growth inhibition and cell death via DNA cleavage. We have attempted to determine if net DNA cleavage is directly related to growth inhibition by measuring whether both parameters vary in parallel. Of primary importance to these studies was use of several bleomycin congeners. We have shown that these congeners vary in their abilities both to inhibit cell growth and to cause DNA damage. Bleomycin B2, tallysomycin, and phleomycin were the most potent growth inhibitors, and bleomycin B2 caused the most DNA damage. N-Acetylbleomycin A2 was inactive in both assays. The net amount of DNA damage measured at two levels of growth inhibition was compared for each congener and was found to vary widely among the congeners. Similarly, the degree of growth inhibition at a given level of submaximal DNA damage was found to vary widely when individual congeners were compared to each other. Hence, growth inhibition and net DNA damage due to bleomycin are not directly correlated with each other when individual congeners are compared to each other.

Reversed-phase paired-ion high-performance liquid chromatographic method for the separation and quantification of multiple bleomycin congeners

A rapid, linear gradient chromatographic technique for separating and quantifying copper(II)-chelated bleomycin congeners is described. This method is also capable of separating divalent from trivalent metal chelates; determining the purity of many chemically modified bleomycins; and assaying bleomycin hydrolase activity on complex mixtures. Quantification at 280 nm is sensitive to 50 pmol and is linear at least up to 10 nmol per injection.

The conformation and orientation of copper (II)-bleomycin intercalated with DNA

EPR data are used to describe the conformation and identity of the atoms coordinated to Cu(II) in Cu(II)-bleomycin bound to oriented DNA fibers. The fibers were slowly drawn from viscous solutions of Cu(II)-bleomycin-DNA containing one Cu(II)-bleomycin to 200 basepairs. EPR measurements were made at room temperature and 90 K for different orientations of the external magnetic field with respect to the helical axes of the fibers. The g-values (g parallel = 2.21, g perpendicular = 2.04) and the hyperfine constant (A parallel = 175 G) are consistent with values expected for Cu(II) chelated to a square planar array of ligands. In the oriented fibers, the square planar arrays do not all have the same orientations with respect to the fiber axes. At room temperature the chelated ions have rotational freedom in which the normal to the planar array has almost complete freedom of rotation about axes perpendicular to the DNA fiber axes. The normal maintains an angle of 75 degrees with respect to the axis, in the plane of the basepair, about which it rotates. Nine superhyperfine peaks on the high field side of the EPR spectrum were partially resolved. The number and splitting (12 G) of these superhyperfine peaks indicate that four nitrogen atoms are chelated to Cu(II) in a square planar array. These data on Cu(II)-bleomycin bound to DNA give information on the orientation of the metal-containing portion of bleomycin which lies outside to double helix.

Demonstration of nitrogen coordination in metal--bleomycin complexes by electron spin--echo envelope spectroscopy

We have studied the Cu(II), Co(II), and Fe(III) complexes of the antineoplastic drug bleomycin by using electron spin--echo envelope spectroscopy. For all three complexes, nitrogen coordination of the metal ions is demonstrated. For the Cu(II)-- and Co(II)--drug complexes, we have been able to identify imidazole as a metal ligand.

Distance geometry analysis of the N.M.R. evidence on the solution conformation of bleomycin

Conformational constraints derived from n.m.r. experiments, X-ray data and the known stereochemistry have been used to investigate by the distance geometry method the range of allowed solution conformations for Cu(II):P-3A (a biosynthetic precursor of bleomycin), Fe(II):bleomycin:carbon monoxide, and Zn(II):bleomycin. The experimental data have been found to be self-consistent and lead to the following observations. 1) Designation of the ligands and the dihedral angles available from vicinal coupling constants are not sufficient to define uniquely the geometry around the metal. 2) When only five bleomycin ligands are invoked (e.g. Cu(II):P-3A or Fe(II):bleomycin:carbon monoxide) there is considerable freedom in the allowed coordination scheme around the metal, but some regions of the molecule have well determined conformation. 3) Introduction of a sixth bleomycin ligand, as in Zn(II):bleomycin, considerably constrains the conformational freedom of the groups coordinated to the zinc. The utility of the distance geometry approach for analysis of data and design of experiments is discussed.