Science and Visualization Science and VisualizationThe polycyclic aromatic hydrocarbons can become chemically altered in our bodies, changing into highly reactive substances called diol epoxides. These diol epoxides then can bind chemically to DNA, forming substances called carcinogen-DNA adducts.
The creation of these carcinogen-DNA adducts is widely believed to be a critical initiating step in the carcinogenic process. The adducts can cause mutations in key genes involved in growth control, which can ultimately result in the formation of tumors.
Remarkably, the diol epoxides that result from the biological activation can have markedly different tumorigenic potentials even when they are chemically very similar. This has been a subject of intense interest among biologists and chemists in the field of cancer research.
Benzo[a]pyrene is a polycyclic aromatic hydrocarbon that has long been studied as a paradigm for the fascinating chemical structure/tumorigenic potency relationship. In particular, benzo[a]pyrene can be biochemically altered into, among others, one pair of diol epoxides that are mirror images of each other; these are known as (+) and (-) anti benzo[a]pyrene diol epoxide (BPDE).
Intriguingly, the (+) BPDE is tumorigenic in rodents, while the (-) BPDE is not. Both link themselves to DNA at the same site, the amino group of the base guanine. For nearly twenty years, a leading question has been how the (+) and (-) BPDE-DNA adducts differ in structure.
A combined multidisciplinary approach has permitted the elucidation of these structures. Breakthroughs in the chemical synthesis of these adducts, pure and in large quantities (milligram amounts) occurred in the group headed by Dr. Nicholas E. Geacintov, professor of chemistry here at New York University, with vital contributions by Dr. Shantu Amin of the American Health Foundation. High resolution nuclear magnetic resonance (NMR) data were obtained and interpreted in the laboratory of Dr. Dinshaw J. Patel, now at Memorial Sloan Kettering Cancer Center. Dr. Monique Cosman, while a graduate student in the Geacintov laboratory and later a post-doctoral researcher in the laboratory of Patel, also played a key role. And finally, the molecular views that agreed with the data were computed by molecular mechanics calculations by Dr. Brian E. Hingerty of the Oak Ridge National Laboratory and Dr. Suse Broyde, professor of biology at NYU.
The results revealed that the BPDE adducted moiety is aligned oppositely along the DNA double helix in the (+) and the (-) case. It is situated in what is known as the minor groove, at the helix exterior. However, the DNA double helix has a directionality, since it is not symmetric. The directionalities are known as the 5-prime direction and the 3-prime direction. In the (+) case the
BPDE moiety points in the 5-prime direction of the damaged strand, while it points 3-prime in the (-) case. In a fixed view of the DNA, this can be described as pointing up or pointing down (see Figure 1 at right: (+) BPDE DNA adduct on left; (-) BPDE DNA adduct on right).
Interestingly, this opposite orientation phenomenon had been predicted from molecular mechanics computations in the Broyde-Hingerty-Geacintov collaboration, prior to the independent synthesis and high resolution NMR experiments.
The finding for this mirror image pair of BPDEs has now proved to be a broad, general principle, true for other (+) and (-) pairs of diol epoxides stemming from different polycyclic aromatic hydrocarbons, whose carcinogenic potentials differ. The diol epoxide may bind to a different base than guanine, notably the base adenine; it may reside in a position different from the minor groove; and the diol epoxides originating from different polycyclic aromatic hydrocarbons have different numbers and types of rings in them. Nonetheless, the members of the (+)/(-) pair are oriented oppositely in the DNA in every case observed so far, some half dozen pairs at least.
The origin of the opposite-orientation effect has been studied extensively by Xiao-ming Xie, a graduate student in the Chemistry Department who is advised by Professors Nicholas E. Geacintov and Suse Broyde. He has carried out extensive molecular mechanics calculations of the BPDE (+)/(-) adduct pair in a very simple DNA sub-unit, a nucleoside, which contains just one BPDE-adducted guanine base together with its attached sugar (see Figure 2 at right: (+) BPDE nucleoside adduct on top; (-) BPDE nucleoside adduct below).
A large number (373,248) of possible structures were created for the (+) case, and a like number for the (-), permitting a very thorough search for all possible structural types. Energies of all the structures were computed, and low-energy, favored structures were evaluated. Four types of low-energy structures were found from these computations for the (+) adduct and four for the (-).
Remarkably, each of the four structures from the (+) adduct was a mirror image of one of the four from the (-) set at the base guanine-adducted BPDE level. Only the sugar attached to the guanine broke the symmetry (Figure 3, below). Moreover, the origin of the opposite-orientation phenomenon became clear: when the BPDE moiety in the (+) case was turned to the position it adopts in the (-), and vice-versa, the structure became too crowded, a phenomenon known as steric hindrance.
Figure 3: The four pairs of structures for (+) and (-) BPDE nucleoside adducts. in each pair, (+) is on the left and (-) is on the right. The guanine is being viewed edge on. Note the mirror-image symmety in the members of the pairs, broken only by the sugar which is at the bottom of the structure.
The striking opposite-orientation effect is a plausible underpinning behind different biological outcomes, since the enzymes that must interact with the lesions during DNA replication and repair would be confronted with opposite orientations, if the opposite-orientation phenomenon were also true under the biological conditions. Thus, this work has opened a door to the possibility of understanding a basic structural reason for differences in carcinogenic potential of chemically very similar, even mirror-image molecules. Ultimately, one hopes to develop a library of structural hallmarks associated with mutagenicity and carcinogenicity. Then, it might be possible to computationally predict which substances are harmful and which benign. This would avoid the laborious, expensive and controversial tests in mammals that are now absolutely necessary to identify carcinogenic substances.
![[ C ]](../icons/CSmSp98.gif){kind=icon}
For more information about ACF scientific computing resources, upcoming events and workshops, or labs and services, please visit the ACF Science and Visualization website at sciweb.nyu.edu.
Posted January 20, 1998
 |
 
|
|
|