Structural
            variations

Amiloride is one of a class of molecules, called intercalators, that bind to DNA. Amiloride likes to bind to stretches of DNA rich in adenine and thymine, two of the four basic building blocks of the DNA molecule. (The others are cytosine and guanine.) The DNA molecule resembles a twisted ladder, with pairs of building blocks, or bases, making up the rungs. Adenine is always paired with thymine, and cytosine is always paired with guanine. Biologists read the code of a gene by determining the sequence of the base pairs along the DNA molecule. The sequences are spelled out in a genetic alphabet consisting of four letters: A, T, C, and G.
  Side view of DNA/amiloride complex
  IMAGE: Dmitry Bondarev (Rutgers University, Newark) and Carol Venanzi (New Jersey Institute of Technology)
The illustration above shows an amiloride molecule bound to a DNA molecule. The DNA, shaped like a twisted ladder, is made up of six pairs of chemical building blocks called bases. All six base pairs are made up of thymine and adenine, two of the four bases found in DNA. The amiloride molecule is composed of six types of atoms: hydrogen (white), nitrogen (blue), carbon (gray), chlorine (lime), oxygen (red), and phosphorous (purple). It is shown nestled between two rungs of the DNA molecule: adenine-thymine (top) and thymine-adenine (bottom). The stick-like structures in the DNA indicate the geometry of the atomic bonds in the bases.


Amiloride and other intercalators latch onto DNA by wedging themselves between two neighboring base pairs. But the puzzle of amiloride is that unlike most other known intercalators, it prefers binding to AT-rich sequences of DNA rather than CG-rich regions. Venanzi saw this as a juicy fundamental problem in molecular recognition. "We wanted to look at different DNA sequences and see if there's something about the structure that would lead to a preference for AT-rich regions," she says. Structural variations might reveal underlying principles of molecular recognition that could, for instance, help in the design of drugs that act at very specific sites, thus reducing unwanted side effects.

So, with a major grant of computing time at NCSA, Venanzi and a graduate student, Dmitry Bondarev, are modeling dozens of different DNA sequences. (Bondarev is a Ph.D. candidate under Venanzi's direction at Rutgers University's Newark campus.) They hope to find properties of the DNA molecule that might help explain why amiloride prefers binding sites with lots of adenine and thymine. To keep down the amount of computation, they're using DNA sequences of six base pairs in length, known as hexamers. These hexamers can be imagined as DNA ladders with only six rungs.



 
1 2 3